Chitin and chitosan: transformations due to the electrospinning process.

INTRODUCTION

The electrospinning process fabricates non-woven fibrous mats, which have smaller fiber diameters and larger surface-to-volume ratios than mats produced by conventional methods of fiber production. The typical electrospinning apparatus includes a high power supply, capillary tube with needle, and target (1). The capillary tube contains a polymeric solution that is held at a constant distance from the target. Upon applying voltage to the needle, which is connected to the power supply, droplets of polymer solution are held at the end of the needle, as a result of surface tension. A Taylor cone (2), which is a conical protrusion, forms (3) when a critical voltage is obtained. From this cone, a straight jet emerges; however, this straight segment lasts for just a few centimeters. Next, the jet exhibits a conical shape, within which exists the complicated path taken by the polymer jet (4). Bending instabilities are experienced by the conically moving jet whose field is directed towards the oppositely charged target. Dry polymer fibers are deposited on the target, since in the short time it takes the jet to reach the target, a majority of the solvent has evaporated (5). Notably, there are a number of parameters, including the properties of the polymer, polymer solution, and processing variables that determine if polymeric solutions will electrospin (6-8).

Because of the intrinsic biological and physicochemical properties of the biopolymers chitin and chitosan, such as biocompatibility and biodegradability (9), (10), nanofibrous mats composed of these materials could be used for wound dressings (11), (12), tissue engineering scaffolds for drug delivery (13-17), or filters for metal recovery (18). Chitin is a nitrogen-rich, high-molecular weight linear polysaccharide composed of N-acetyl-D-glucosamine (N-acetyl-2-amino-2-deoxy-D-glucopryanosegiucopryanose) units linked by [beta]-D (l [right arrow]4) bonds. Chitin can be challenging to work with due to its insolubility in most organic solvents. The biopolymer is soluble in hexafluoroacetone, hexafluoroiso-propanol (HFIP), chloroalcohols in conjunction with aqueous solutions of mineral acids, and dimethylacetamide containing 5% lithium chloride (19), (20). Chitosan, the N-deacetylated derivative of chitin, is commercially ~85% deacetylated (19). The free amines allow chitosan to be soluble in aqueous acidic solvents that chitin cannot dissolve in, such as formic acid, acetic acid, and malic acid. This change is a result of the protonation of the--[NH.sub.2] function on the C-2 of the D-glucosamine repeat unit, and creates a polyelectrolyte in acidic solutions (9), (19).

Despite the challenges of working with the biopolymers chitin (21) and chitosan (22), they both have previously been electrospun. The appropriate solvent systems (18), (23), methods of neutralization (24) and cross-linking (25) for chitosan have also been investigated. Additionally, although experimental research in combination with mathematical modeling has explained the complexities of the electrospinning process (4), (26-28), no investigations have focused on the polymer throughout the course of the electrospinning process. Molecular structure and the functionality of chitosan can be investigated by X-ray diffraction (XRD) (29). There are a number of relevant lower angle crystalline reflections commonly reported for chitin and chitosan between 5[degrees] and 27[degrees], they are (020), (110), (120), and (130) (30), (31). Additionally, a peak at (040) is considered as mixed (30, 31) with the (110) peak because they are close together but the (040) is much weaker (30). The investigation of chitin and chitosan, which are neutral and polyelectrolytic biopolymers, respectively, provides an understanding as to the range of properties found in semi-crystalline polymers that can be electrospun. The use of these biopolymer fibrous meshes is dictated by their stability and crystallinity.

We have previously electrospun practical grade (PG) chitin (8), medium molecular weight (MMW) chitosan, and PG chitosan (32). Sangsanoh and Supaphol (24) have reported that supersaturated solutions of sodium carbonate can be utilized to neutralize the solvent required for spinning the chitosan mats. Additionally, glutaraldehyde (GA) has been demonstrated to improve the chemical integrity of electrospun chitosan fibrous mats through cross-linking using either a one-step (25) or two-step (32) process. The chemical stability due to neutralization and cross-linking, as well as the effects that electrospinning have on the crystalline integrity of the as-spun chitin, chitosan, and cross-linked chitosans will be investigated by field emission scanning electron microscopy (FESEM), solubility testing, and XRD.

EXPERIMENTAL

Materials

All compounds were used as received. Sodium carbonate monohydrate granular ([Na.sub.2][CO.sub.3]) was purchased from Mallinckrodt (Paris, Kentucky). Fifty wt% GA in water, 97% pure sodium hydroxide (NaOH), [99.7.sup.+] ACS reagent-grade acetic acid (AA), ReagentPlus 99% trifluoroacetic acid (TFA), 1,1,1,3,3,3-hexafluoro-2-propanol 99+% (HFIP), PG chitin from crab shells [degree of deacetyla-tion (DD): 9%], PG chitosan from crab shells [molecular weight (MW): ~190,000-375,000 and DD: 83%], and MMW chitosan (MW: ~190,000-310,000 and DD: 75%) were purchased from Sigma-Aldrich (St Louis, MO). Room temperature, ultrapure water was used to make the solubility test solutions.

Solution Preparation

Solutions of 3.1% (w/v) chitin/HFIP and 2.7% (w/v) chitosan/TFA were mixed for at least 72 h and 24 h, respectively, on an Arma-Rotator A-1 (Bethesda, MD). For one-step cross-linked solutions, 1 mL of GA was added to 15 mL of 2.7% (w/v) chitosan/TFA solution and dispersed for 5 s using a Vortex-Genie (Scientific Industries, Bohemia, NY).

Electrospinning

The electrospinning apparatus utilized was previously described (32). A 5 mL Luer-lock syringe was loaded with solution (Becton Dickinson, Franklin Lakes, NJ) and a Precision Glide 21-gauge needle (Becton Dickinson, Franklin Lakes, NJ) was attached. The positive electrode of a high-voltage supply (Gamma High Voltage Research, Ormond Beach, FL) was directly connected to the needle by an alligator clip. The syringe was then placed on an advancement pump (Harvard Apparatus, Plymouth Meeting, PA) located a fixed distance from the negative electrode. Voltage was then applied, thus creating a positive and negative anode as the solution was advanced at a constant rate. To electrospin chitin, an advancement speed of 1.2 mL/h, a separation distance of 6.0 cm, and an applied voltage of 24 kV were used. Chitosan solutions were electrospun using an advancement speed of 1.2 mL/h, a separation distance of 6.4 cm, and ~26 kV of applied voltage. For all experiments, a copper plate wrapped in aluminum foil was utilized as the collector. The temperature and relative humidity (%) in the laboratory during electrospinning were monitored by a digital thermohygrometer (Fisher Scientific, Pittsburgh, PA).

One-Step Cross-Linking

The chitosan/TFA/GA solutions as previously reported (25) were electrospun using the same parameters as utilized for the chitosan/TFA solutions.

Two-Step Cross-Linking

Electrospun fibrous mats of chitosan were placed into a 11.43 X 7.62 X 5.08 [cm.sup.3] vaporization chamber (VWR Scientific Products, Bridgeport, NJ) for 24 h as previously reported (32). On the bottom of the chamber was 3 mL of aqueous GA solution, which vaporized at room temperature (25[degrees]C).

Neutralization

As-spun MMW chitosan mats were neutralized following the method of Sangsanoh and Supaphol (24). The mats were submerged in 5 M aqueous saturated solutions of [Na.sub.2][CO.sub.3] for 3 h at room temperature (25[degrees]C), where there was an excess amount of undissolved [Na.sub.2][CO.sub.3] present. DI water was utilized to rinse the mats to obtain a neutral pH where upon they were desiccated for at least 24 h.

Field Emission Scanning Electron Microscopy

Micrographs of the electrospun fiber mats were obtained with a Zeiss Supra 50/VP FESEM. A Denton vacuum desk II sputtering machine was utilized to coat the samples for 5-10 s with platinum-palladium.

Solubility Testing

The solubility of the chitin and neutralized MMW chitosan fibrous mats were tested in the same manner as previously tested chitosan and cross-linked chitosan fibrous mats (32). Three 15-[mm.sup.2] petri dishes (Becton Dickinson, Franklin Lakes, NJ) contained 30 mL of three various solutions: basic (1 M NaOH), acidic (1 M AA), and aqueous (ultrapure [H.sub.2]O). Two samples of chitin fibrous mats, each 2.54 X 1.27 cm were placed into each solution. After 15 min, if possible, one of the mats was removed, whereas the other remained in the solution for 72 h.

X-Ray Diffraction

XRD Patterns were recorded using a D500 Siemens XRD with a CuK[alpha] source. Scans taken of bulk materials consisted of ground biopolymer sample, which was compacted onto glass slides. Films of chitin/HFIP and chitosan/TFA solutions and all fiber scans were mounted onto glass slides wrapped with store purchased aluminum (Al) foil. All X-ray diffractograms are graphed with 2[theta] values ranging from 5 to 28[degrees] and were acquired using a hold time of 0.04 s. The data output files were further analyzed for peak detection, as well as subtraction of Al foil peaks, when necessary, using the MDI JADE 7 software. Crystallinity index (CI), as previously applied to chitin and chitosan (33-35) and before that proposed for use for cellulose (36) is given by the following equation:

Crystallinity Index(CI)(%) = [([I.sub.110] - [I.sub.am])/[I.sub.110]] X 100 (1)

where: [I.sub.110] (arbitrary units) is the maximum intensity of the (110) peak, which is usually around 2[theta] = 19[degrees] and [I.sub.am] (arbitrary units), which is the amorphous diffraction at 2[theta] = 12.6[degrees]. We slightly modified the given mathematical equation. Instead of using a single data point, we averaged three points, one intensity measurement to the left of, and one to the right of [I.sub.110] and [I.sub.am]. For example, the [I.sub.am] points were always taken at 2[theta] = 12.56[degrees], 12.60[degrees], and 12.64[degrees].

RESULTS AND DISCUSSION

Chemical Integrity of Electrospun Chitin and Chitosan Mats

Previously, PG chitin, dissolved in HFIP (8) was electrospun into fine, cylindrical, continuous, bead-free, randomly oriented fibrous mats [Refer Appendix for Fourier transform infrared spectroscopy (FTIR) on bulk and electrospun chitin]. The as-spun chitin mats were subjected to 1 M acetic acid (AA), 1 M sodium hydroxide (NaOH), and ultrapure water ([H.sub.2]O) solutions for a 15 min and 72 h to determine their chemical stability. The term "as-spun" indicates that no cross-linking or other post-processing has occurred to the electrospun fibers. Table 1 describes the results of the 72 h solubility tests, a "[check]" indicates that visually, the electrospun mat appears to have survived the solution and could be removed. Whereas an "X" means that the mat could not be removed from solution, or appeared to have lost its fibrous form. The results of the 15 min test were the same as the 72 h; if a mat could not be removed from the solution, it disintegrated or dispersed instantaneously into the solution upon contact. The chitin mats survived all tests. After being removed from all aqueous solutions, visually, the mats retained their rectangular shape and color. Field emission scanning electron microscope (FESEM) imaging, as displayed in Fig. 1, provides supporting evidence confirming that a porous fibrous structure is retained within the mats. Table 2 displays that after removing the chitin fibrous mats from the solutions, their average fiber diameter increased; however, they all remained within three standard deviations of the initial as-spun diameter. Chitin is insoluble in most common organic solvents whereas chitosan dissolves in weakly acidic solutions, hence, it is necessary to cross-link mats composed of chitosan. Two cross-linking methods have already been identified. In the one-step (25) cross-linking method, fibers are cross-linked with GA-liquid in situ during the electrospinning process. Whereas, for production of the two-step (32) cross-linked fibers, first the fibers are electrospun as a mat and then, the mat is subjected to a vaporization chamber that contains GA-vapor. After those two steps are completed, the resultant fibrous mats are cross-linked. Previously, we have reported, based on visual inspection that one-step and two-step cross-linked chitosan fibrous mats have increased chemical stability over as-spun mats (refer Table 1) (25), (32). For the first time, SEM imaging (refer Fig. 2) has been conducted in an effort to confirm these results and to determine the change in average fiber diameter (Table 2). Figure 2c and d display the two-step cross-linked MMW chitosan fibers which retain their cylindrical integrity when subjected to basic and aqueous solutions. While many MMW chitosan fibers retained their morphology when submersion in acidic solutions (Fig. 2a), other fibers were found to degrade slightly, refer Fig. 2b. The overall fibrous structure was maintained, however, the fibers that experienced degradation somewhat lost their cylindrical shape.

[FIGURE 1 OMITTED]

[FIGURE 2 OMITTED]

TABLE 1. Properties of various electrospun chitin and chitosan fibrous
mats including (1) their fiber diameter before solubility testing and
(2) survival post-immersion in acidic, basic, and aqueous solutions
based on visual inspection.

                       Average fiber         Survival post 72 h
                       diameter (nm)

                                           AA     [H.sub.2]O    NaOH

As-spun PG chitin      152 [+ or -] 70   [check]    [check]    [check]

As-spun PG chitosan     58 [+ or -] 20      X          X       [check]

As-spun MMW chitosan    77 [+ or -] 29      X          X       [check]

One-step cross-linked  128 [+ or -] 40   [check]    [check]    [check]
MMW chitosan

Two-step cross-linked  172 [+ or -] 75   [check]    [check]    [check]
MMW chitosan

Neutralized and        250 [+ or -] 120     X       [check]    [check]
two-step cross-linked
MMW chitosan

A "[check]" indicates that visually, the electrospun mat survived and
could be removed from the solution. An "X" indicates that the mat could
not be removed from solution, or appeared to have lost its fibrous
form.

TABLE 2. Average fiber diameter of MMW chitosan electrospun mats which
have been two-step cross-linked with and without neutralization.

                              Average fiber diameter (nm)
                               Post 72 h immersion in:

                             AA          [H.sub.2]O        NaOH

As-spun chitin         315[+ or -]136  290[+ or -]100  306[+ or -]99

Two-step cross-linked  395[+ or -]132  346[+ or -]145  295[+ or -]124
MMW chitosan

Neutralized and        N/A             450[+ or -]150  380[+ or -]165
two-step cross-linked
MMW chitosan

Sangsanoh and Supaphol (24) have previously demonstrated using sodium carbonate solutions to remove remnant salt present due to the unevaporated solvent, trifluoroacetic acid (TFA). After this process, they reported that the mats retained their fibrous structure post 12 weeks in pH 7.4 pH buffer saline solution or distilled water. In an effort to better investigate the physical and chemical changes occurring as a result of two-step cross-linking, we first neutralized our chitosan mats and then two-step cross-linked them (refer FESEM images in Fig. 3a and b, respectively). FTIR confirmation of these steps can be found in the Appendix. After 72 h in acidic solutions, the fibrous mats were removed and visually appeared to be fibrous mats. However, the micrograph, Fig. 3c, reveals that these mats lost their fibrous structures. Fibrous structure was retained when the mats were immersed in basic and aqueous solutions (Fig. 3d and e). Fiber diameter changes due to these processes are displayed on Tables 1 and 2. The average fiber diameter increased significantly, greater than three standard deviations from the initial measurements of the as-spun MMW chitosan to the time they were removed from the solubility testing post-neutralization.

[FIGURE 3 OMITTED]

Crystallinity Analysis of As-Spun Fibrous Mats

For the best XRD analysis, proper sample preparation is imperative; the sample must contain enough particles mounted with all orientation angles exposed so that diffraction of all possible diffracting planes will occur (37). Because of the small fiber diameter of electrospun fibers, it is thought that XRD could provide reliable data since all orientations of the electrospun semi-crystalline polymer should be present. Reproducibility of acquiring XRD data on nanofibrous mats mounted on Al foil was our first concern. The standard error of the instrument, the standard error of the sample (sample variability), and a normal range of standard deviation were determined (refer Appendix). Thus, we used this technique to compare the diffraction patterns of the bulk biopolymers to their electrospun mats. Figure 4 contains the diffractograms of bulk chitin, filtered solutions of chitin dissolved in HFTP, and electrospun chitin. Since chitin is only ~65% (21) soluble in HFIP, the solution was filtered to avoid diffraction peaks from undissolved bulk material. Additionally, the CI (calculated by Eq. 1), and the various d-spacings along with their 2[theta] locations (as determined by the JADE 7 software) are displayed in Table 3.

[FIGURE 4 OMITTED]

TABLE 3. Interpretations of XRD data: crystallinity index and
structural parameters of chitin and chitosans at different points
during the electrospinning process.

                                                    (020)

             Biopolymer             CI (%)   2[theta]  d([Angstrom])

Bulk         Chitin                 94.95       9.3         9.5

             PG chitosan            93.78      10.5         8.4

             MMW chitosan           94.63      10.7         8.3

Solution     Chitin/HFIP

             PG chitosan/TFA

             MMW chitosan/TFA

Fibrous mat  Chilin (A1)            96.69       9.5         9.4

             PG chitosan            92.23

             Two-step PG chitosan   89.71 *

             One-step PG chitosan   90.14 *

             MMW chitosan           87.86 *

             Two-step MMW chitosan  88.80 *

             One-step MMW chitosan  90.91 *

                            Unlabeled                  (110)

          Biopolymer   2[theta]  d([Angstrom])  2[theta]  d([Angstrom])

Bulk      Chitin         14.0         6.3         17.0         5.2

          PG             14.0         6.3         17.0         5.2
          chitosan

          MMW            14.2         6.3         17.0         5.2
          chitosan

Solution  Chitin/                                 17.4 *       5.1 *
          HFIP

          PG                                       21.4 *      4.1 *
          chitosan/
          TFA

          MMW                                      20.3 *      4.4 *
          chitosan/
          TFA

Fibrous   Chilin         14.3         6.2          17.0 *      5.2 *
mat       (A1)

          PG             14.0         6.3          16.9 *      5.2 *
          chitosan

          Two-step PG    14.1         6.3          17.0 *      5.2 *
          chitosan

          One-step PG    14.2         6.2          17.1 *      5.2 *
          chitosan

          MMW            14.3         6.2          17.1 *      5.2 *
          chitosan

          Two-step       14.1         6.3          17.0 *      5.2 *
          MMW
          chitosan

          One-step       14.4         6.2          17.1 *      5.2 *
          MMW
          chitosan

                              (120)                    (101)

          Biopolymer   2[theta]  d([Angstrom])  2[theta]  d([Angstrom])

Bulk      Chitin         18.4         4.8

          PG             18.5         4.8          19.8        4.5
          chitosan

          MMW            18.7         4.8          20.7        4.4
          chitosan

Solution  Chitin/
          HFIP

          PG
          chitosan/
          TFA

          MMW
          chitosan/
          TFA

Fibrous   Chilin         18.8         4.8
mat       (A1)

          PG             18.4         4.8
          chitosan

          Two-step PG    18.6 *       4.8 *
          chitosan

          One-step PG    18.8         4.7
          chitosan

          MMW            18.8         4.7
          chitosan

          Two-step       18.6         4.8
          MMW
          chitosan

          One-step       18.8         4.7
          MMW
          chitosan

                                           (130)

             Biopolymer             2[theta]  d ([Angstrom])

Bulk         Chitin                   25.6          3.5

             PG chitosan              25.6          3.5

             MMW chitosan             25.7          3.5

Solution     Chitin/HFIP

             PG chitosan/TFA

             MMW chitosan/TFA

Fibrous mat  Chilin (A1)              25.8 *        3.5 *

             PG chitosan              25.6          3.5

             Two-step PG chitosan     25.7 *        3.5 *

             One-step PG chitosan     25.8          3.5

             MMW chitosan             25.8          3.5

             Two-step MMW chitosan    25.6          3.5

             One-step MMW chitosan    25.7          3.5

CI can not be measured for the dissolved polymer since the 110 peak was
not displayed.

* Indicates that a statistically significant change has occurred in
comparison to the bulk material.

The diffractograms taken on bulk chitin and chitosan (Figs. 4a, 5a, and 6a) were compared with their diffractograms taken of the biopolymers dissolved in solution (Figs. 4c, 5c, and 6c). It was determined that there is a statistically significant decrease in the CI, which was expected. It has previously been reported that dissolution of chitosan in AA causes a decrease in its crystallinity (34). After dissolution of the biopolymers, notably, the (110) peak of chitin and as-spun MMW chitosan fibers broadens significantly indicating the solution is amorphous. It should be noted that the "dip" that appears on the diffraction pattern taken on the solution scan is a remnant from subtracting the Al foil peaks and does not interfere with any of the measurements.

[FIGURE 5 OMITTED]

[FIGURE 6 OMITTED]

Wada and Saito (30) determined that [alpha]-chitin is structurally stable; upon heating the biopolymer from room temperature to 250[degrees]C, it did not experience a thermal decomposition or a crystalline phase transformation. It was of interest in this investigation to determine if chitin is also structurally stable after being dissolved in HFIP and the application of relatively high voltage. This is likely as electrospinning the amorphous solution into non-cross-linked fibrous mats resulted in an increased amount of crystallinity. These mats regain much of their original crystallinity despite the speed in which the entire electrospinning process occurs. Many of the reported d-spacings of the electrospun mats do not experience statistically significant changes when compared to the bulk. Two d-spacings do statistically decrease, shift to higher 2[theta] angles, for chitin they are the (110) and (130) peaks. The (130) has previously been reported as being indicative of orientation (38). The (130) peak of PG chitosan also decreases. Upon visual inspection, the unlabeled peak, as well as the (120) peak, broaden and become less sharp, indicating a decrease in the crystallinity of chitin.

Park and coworkers (21) have demonstrated that when electrospun chitin mats are deacetylated into chitosan mats, the peak observed at 2[theta] = 9.2[degrees] decreases. As displayed in Table 3, we as well observe this peak for chitin whereas it is not a statistically significant peak for chitosan. This change results because chitosan has randomly oriented bulky acetyl groups and hence has a less stereo-regular structure than chitin (39).

It is interesting to note that Muzzarelli et al. (40) have reported the lack of a strong (020) peak for chitin. However, they observed a significant peak at d = 7.044. We as well report a significant peak termed "unlabeled" at d = 6.320 [Angstrom] (bulk) and 6.200 [Angstrom] (electrospun). Muzzarelli et al. note that other references (41), (42) do not report this peak as observable.

Figures 4b and 5b indicate that visually, the peaks of as-spun fibrous mat of PG and MMW chitosan are less sharp than the bulk materials. Upon recrystallization, all electrospun chitosan fibers loose the (101) peak around 20[degrees]. This peak has been reported to decrease as a result of shrinkage of cell size and decreasing intersegmental spacing. This loss in crystallinity is said to improve the usage of the fibrous mats for filtration and membrane applications (43).

Crystallinity Analysis of Cross-Linked Fibrous Mats

After the diffractograms were taken of the as-spun PG and MMW chitosan mats, the same samples were two-step cross-linked with vapor-GA for 24 h and their dif fraction patterns collected. Additionally, one-step liquid-GA cross-linked PG and MMW chitosan fibrous mats were fabricated. Refer Figs. 5 and 6, as well as Table 3.

The (101) peak, which as previously mentioned was not significant enough to be recognized by the JADE soft ware, has been reported to decrease with cross-linking density due to the changes that the--[NH.sub.2] and--OH functional groups undergo due to chemical cross-linking based on decreases in the crystallinity of chitosan (43). In addition to the rationales listed in the previous section, the application of GA could account for the loss of this peak in the one-step electrospun fibers.

Homogeneous cross-linking (44) of chitosan with GA decreases the crystallinity of chitosan, while heterogeneous cross-linking (45) does not affect the crystallinity. Also, the crystallinity of chitosan has been demonstrated to decrease with increasing amounts of GA (43), (46) due to the decreased cell size. The one-step cross-linking method homogeneously cross-link the fibers and two-step cross-linking method heterogeneously cross-link the fibers, however, details regarding the cross-linking, for instance the extent of cross-linking, remain unclear. The DD of the chitosan effects the ability to form covalent GA cross-linking (47) and it is likely that the application of voltage changes the DD of the chitosan. On the basis of these factors, we expect cross-linking not to have a major effect on crystallinity, but to change the observed peak locations and intensities.

When comparing the bulk materials to the as-spun and cross-linked fibrous mats, almost all mats were determined to have experienced a statistically significant decrease in CI. It was determined that all cross-linked mats (110) d-spacing statistically decreased; perhaps the GA-vapor is contracting the crystal lattice. The one-step PG chitosan fibrous mats experienced another statistically significant change, their (120) and (130) d-spacing decreased. This change in the (130) peak is consistent with the change noted for the electrospun chitin sample. Possibly, the addition of TFA and GA disrupts the sheet formation of chitosan chains.

In visually comparing the diffractograms of the as-spun chitosans in Figs. 5 and 6 to the cross-linked fibrous mats, the cross-linked mats' peaks are less sharp. Notably, the one-step cross-linked fibers (Fig. 5d and 6d) display the least sharp peaks out of all of the diffractograms taken. This higher loss of crystallinity experienced by one-step cross-linking over as-spun and two-step cross-linked fibers is expected. The electrospinning process happens very quickly, and the introduction of TFA/GA interactions is interrupting the recrystallization of chitosan. Notably, there are a variety of reaction times reported for the reaction between chitosan and GA. Knaul et al. (48) noted in 5 min at 25.8[degrees]C and Groboillot et al. (49) observed that in 3 min suspended chitosan microcapsules in organic solutions containing GA were cross-linked. Montiero and Airoldi (46) reported longer cross-linking times, that chi-tosan/AA solutions fully reacted with GA within 1 h. In electrospinning solutions containing GA, there is a maximum amount of time that the solutions can stand before becoming cross-linked. Additionally, allowing the two-step fibers to cross-link in a vaporization chamber for 24 h is longer than necessary. Future investigations will examine the minimum reaction time needed for chitosan to be cross-linked by GA for both the one-step and two-step processes.

CONCLUSION

All cross-linked chitosan fibrous mats as well as chitin fibrous mats survived for 72 h in aqueous and basic solutions. Although some degradation occurred to the fibrous mats when in acidic solutions, degradation of the fibrous mats is desirable for some biomedical and environmental applications. A new cross-linking method for neutralized fibrous mats should be investigated for applications that require a fibrous structure in acidic environments.

The repeatability of conducting XRD analysis on nanofibrous mats was conducted and standard deviations on various common chitin peaks were generated. After electrospinning, the diffractograms of chitin, PG chitosan, and MMW chitosan exhibit peaks that are similar to the bulk materials but broader, indicating the loss of some crystallinity. The one-step cross-linked chitosan fibers might have the lowest crystallinity of all electrospun fibers based on their diffractograms.

Since as-spun chitin, one-step cross-linked chitosan, and two-step cross-linked chitosan mats are all chemically stable fibrous mats composed of non-toxic, renewable materials, understanding the differences in their crystallinity determines their usefulness in a variety of applications. The chain packing and DD determine the kinetics and sorption equilibrium of water and metal ion's ability to reach the sorption sites (34). The application of voltage during the electrospinning process, along with the necessity to cross-link chitosan fibrous mats using the free amine groups directly influences these sites. Some amount of amorphous regions are desirable, as they are more permeable to aqueous solutions than crystalline regions (50); therefore, the necessary amount of fluid flow for a particular application must be known. We foresee electrospun chitin and chitosan utilized for filtration or tissue engineering applications and engineering their crystallinities will allow for enough solution to pass into the system to be tested, or allow the appropriate amount of nutrients to reach the cells.

ACKNOWLEDGMENTS

J.D.S. thank the National Science Foundation (NSF)-Integrative Graduate Education and Research Traineeship (NSF IGERT) and Graduate Assistance in Areas of National Need-Drexel Research and Education in Advanced Materials (GAANN-DREAM), which is funded by the Department of Education's Office of Postsecondary Education for funding. L.A.S. thanks NSF for REU support.

APPENDIX

Fourier Transform Infrared Spectroscopy

FTIR spectra for the bulk biopolymers and electrospun chitin and chitosan fibrous mats were measured using transmission mode. All spectra were taken in the spectral range of 4000-500 [cm.sub.-1] by accumulation of 64 scans and with a resolution of 4 [cm.sub.-1]. The DD of chitosans was previously determined (32), whereas the DD of chitin was calculated from the FTIR spectrum using the following equation (51), (52):

Degree of Deacetylation (DD)(%) = 100 - [([A.sub.1655]/[A.sub.3450]) X 115] (2)

Chitin FTIR Results

Figure A1 displays the FTIR spectra of both the bulk and electrospun PG chitin, which, as expected, are quite similar. With lower DD, chitin exhibits two peaks in the mid-3000 [cm.sub.-1] region indicative of the statistical mixing of [CH.sub.2]OH conformations. Half of these originate from intermolecular hydrogen bonds with OH-6 groups on the neighboring chains close to the gauche-trans (gt) conformation, while the rest are involved with intermolecular hydrogen bonds with C = O groups on the next residue along the chain, close to the trans-gauche (tg) conformation (53), (54). We observe the two groupings of [CH.sub.2]OH groups at 3443 and 3269 [cm.sub.-1] in the bulk and at 3439 and 3300 [cm.sub.-1] in the electrospun fibers. In these peaks, as well as in the C-H stretching bands at 2928 and 2886 [cm.sub.-1] (bulk) and 2929 and 2881 [cm.sub.-1] (fibers), we observe the same phenomena. The second peak, located at the lower wave number becomes less sharp and decreases in intensity post-spinning. This change has been proposed to indicate a change in crystallization (53) due to the electrospinning process.

[FIGURE A1 OMITTED]

For all [alpha]-chitin, such as the PG chitin used, the amide I band, which indicates in-plane NH bending combined with C = O stretching vibrations and CN stretching modes (55) is more conformationally sensitive than the amide II and III vibrations. The amide I usually appears as a split band at 1656 and 1621 [cm.sub.-1] representing the stretching of both the C = O groups hydrogen bonded to the NH and OH-6 groups and the C = O groups hydrogen bonded to NH groups of the neighboring side chain (56). In the bulk spectrum, it appears as peak at 1660 [cm.sub.-1] with a shoulder at 1629 [cm.sub.-1] (9). The electrospun sample does not exhibit a shoulder or a split peak, a single strong peak appears at 1658 [cm.sub.-1]. These changes indicate changes in hydrogen bonding due to the dissolution of chitin in HFIP. During dissolution, the tertiary bonds break down and the hydrogen bonds become disordered (36). Also, the amine II peak, commonly at 1556 [cm.sub.-1] is observed at 1560 and 1556 [cm.sub.-1] for the bulk and electrospun fibers, respectively.

In a previous study by Szabo et al., peptides dissolved in 90/10 HFIP/[D.sub.2]O displayed a peak centered at 1667 [cm.sub.-1], which is close to that of pure HFIP most likely due to peptide's solvated carbonyl groups (57). Upon our thorough comparison of the bulk and as-spun chitin, the loss of a shoulder at 1629 [cm.sub.-1] on the 1661 [cm.sub.-1] peak occurred and the peak recentered at 1659 [cm.sub.-1]. It is theorized that this change is due to the presence of remnant HFIP in the fibers.

FTIR analysis of bulk MMW and PG chitosan as well as their respective electrospun fibers and cross-linked fibers have previously been reported (25), (32). Notably, since chitosan readily forms amine salts (58), the remainder of the solvent used to electrospin the biopolymer is evident by the formation of three bold peaks around 840-720 [cm.sub.-1] and a large adsorption peak at 1675 [cm.sub.-1] or by the presence of a carboxylic acid peak at 1750 [cm.sub.-1] (24), (25), (32).

Additionally, it was determined that both one-step and two-step cross-linked fibers experienced a Schiff base imine functionality as evident by the change in the carbonyl-amide region at 1650 [cm.sub.-1] and the new C = N imine at 1560 [cm.sub.-1] (Refer Fig. A2). The amount of remnant free amine groups will influence the crystallinity of the fibrous mats (34). Differences were noted in the primary and secondary alcohol region between the one-step and two-step electrospun chitosan fibers. Possibly, the differences arise since the glutaraldhyde (GA) and trifluoroacetic acid (TFA) act as a single solvent system in the one-step process, whereas in the two-step process, a solvent substitution occurs since the as-spun fibers contain only TFA until cross-linking, which introduces the GA. Because of the differences in the FTIR of the electrospun cross-linked chitosan fibers, it was suspected that the XRD patterns of the fibers might also exhibit some key differences.

[FIGURE A2 OMITTED]

General Considerations for Analyzing Electrospun Polymeric Fibrous Mats with XRD

By measuring the same sample three times and gathering the instruments standard error, we hoped to investigate the feasibility of using XRD to analyze fibrous mats. The next step was to measure three identical samples to gather the standard error within a sample. One relatively large mat of PG chitin was electrospun onto aluminum (Al) foil and mounted onto glass slides as appropriate for X-ray studies. The chitin samples were labeled A, B1, and C1. The diffraction pattern of chitin A was taken three times and labeled as Al, A2, and A3. Between each run of chitin A, it was removed and the diffraction pattern of a different sample was taken. The obtained diffraction patterns, which are nearly identical, are displayed in Fig. A3.

[FIGURE A3 OMITTED]

Table A1 contains the CI, observed peaks along with their respective 2[theta] locations and d-spacings for those chitin diffractograms. There will be some variation both within one sample (standard error of the instrument), as well as between different samples (standard error of the sample or sample variability). This is evident by the d-spacings reported for the (020) peak on Table A. Consequently, it was important to determine a normal range for the standard deviation.

TABLE A1. Diffraction data acquired on multiple samples of electrospun
chitin including crystallinity index and structural parameters.

                           (020)                  Unlabeled

Sample     CI (%)  2[theta]  d ([Angstrom])  2[theta]   d ([Angstrom])

Chitin A1  96.69     9.4           9.4         14.3           6.2
Chitin A2  96.94     9.0           9.8         14.2           6.2
Chitin A3  97.16     9.6           9.2         14.2           6.2
Chitin B1  95.85     9.3           9.5         14.3           6.2
Chitin C1  97.14     9.0           9.9         14.1           6.3

                   (110)                     (120)

Sample     2[theta]  d ([Angstrom])  2[theta]  d ([Angstrom])

Chitin A1    17.0          5.2         18.8          4.7
Chitin A2    17.0          5.2         18.8          4.7
Chitin A3    17.0          5.2         18.7          4.7
Chitin B1    17.0          5.2         18.8          4.7
Chitin C1    17.0          5.2         18.6          4.8

                   (130)

Sample     2[theta]  d ([Angstrom])

Chitin A1    25.8          3.5
Chitin A2    25.7          3.5
Chitin A3    25.7          3.5
Chitin B1    25.8          3.5
Chitin C1    25.7          3.5

To determine the standard error of the instrument, the statistical repeatability of a single sample was found by scanning chitin sample A three times. The average CI, was 96.93% with a standard deviation ([sigma]) of 0.24. Hence, data becomes statistically different once values lie outside of the average CI [+ or -] 3[sigma], or 97 [+ or -] 0.71%, which gives a range of 96.22-97.64%. The repeatability between three different chitin samples was taken using chitin samples A3, Bl, and C1. The average CI [+ or -] 3[sigma] was 97 [+ or -] 2.3%, which gives a range of 95.97-97.47%. The CI of all chitin samples should fall within this range. However, chitin B1 is outside the 3[sigma] value (by 0.12%) providing the limitations to this kind of analysis for nanofibrous mat analysis.

The variation between the peaks reported on Table A1 was determined to aide in the analysis of the sections to follow. The average d-spacing [+ or -] 3[sigma], as found between a single sample (chitin Al, A2, and A3) and different samples (chitin A3, Bl, and C1) were found to be 9.46 [+ or -] 1.0 [Angstrom] and 9.53 [+ or -] 1.0 [Angstrom] for (020), 6.21 [+ or -] 0.03 [Angstrom] and 6.22 [+ or -] 0.1 [Angstrom] for the unlabelled, 5.20 [+ or -] 0.0 [Angstrom] and 5.20 [+ or -] 0.002 [Angstrom] for (110), 4.73 [+ or -] 0.02 [Angstrom] and 4.74 [+ or -] 0.08 [Angstrom] for (120), and 3.46 [+ or -] 0.02 [Angstrom] and 3.46 [+ or -] 0.02 [Angstrom] for (130) peaks. Notably, the largest peak had the same average d-spacing and the lowest 99.73% confidence interval (as indicated by 3[sigma]), this finding is encouraging.

REFERENCES

(1.) Z.M. Huang, Y.Z. Zhang, M. Kotaki, and S. Ramakrishna, Compos. Sci. Technol, 63(15), 2223 (2003).

(2.) G. Taylor, Proc. R. Soc. London Ser. A, 280(1382), 383 (1964).

(3.) Y.M. Shin, M.M. Hohman, M.P. Brenner, and G.C. Rutledge, Polymer, 42(25), 09955 (2001).

(4.) D.H. Reneker, A.L. Yarin, H. Fong, and S. Koombhongse, J. Appl. Phys., 87(9), 4531 (2000).

(5.) T. Sabbiah, G.S. Bhat, R.W. Tock, S. Parameswaran, and S.S. Ramkumar, J. Appl. Polym. Sci., 96(2), 557 (2005).

(6.) C. Burger, B.S. Hsiao, and B. Chu, Annu. Rev. Mater. Res., 36, 333 (2006).

(7.) A. Greiner and J.H. Wendorff, Angew. Chem. Int. Ed., 46(30), 5670 (2007).

(8.) J.D. Schiffman and C.L. Schauer, Polym. Rev., 48(2), 317 (2008).

(9.) M. Rinaudo, Prog. Polym. Sci., 31(7), 603 (2006).

(10.) B. Krajewska, Enzyme Microb. Technol., 35(2-3), 126 (2004).

(11.) M.S. Khil, D.I. Cha, H.Y. Kim, I.S. Kim, and N. Bhattarai, J. Biomed. Mater. Res. B Appl. Biomater., 67B(2), 675 (2003).

(12.) S.H. Kim, Y.S. Nam, T.S. Lee, and W.H. Park, The Society of Polymer Science. Japan, 35, 185 (2003).

(13.) W.J. Li, C.T. Laurencin, E.J. Caterson, R.S. Tuan, and F.K. Ko, J. Biomed. Mater. Res., 60(4), 613 (2002).

(14.) J. Zeng, X. Xu, X. Chen, Q. Liang, X. Bian, L. Yang, and X. Jing, J. Control. Release, 92(3), 227 (2003).

(15.) G. Verreck, I. Chun, J. Rosenblatt, J. Peeters, A.V. Dijck, J. Mensch, M. Noppe, and M.E. Brewster, J. Control. Release, 92(3), 349 (2003).

(16.) C. Li, C. Vepari, H.-J. Jin, H.J. Kim, and D.L. Kaplan, Biomaterials, 27(16), 3115 (2006).

(17.) Y.K. Luu, K. Kim, B.S. Hsiao, B. Chu, and M. Hadjiargyrou, J. Control. Release, 89(2), 341 (2003).

(18.) S.D. Vrieze, P. Westbroek, T.V. Camp, and L.V. Langenhove, J. Mater. Sci., 42, 8029 (2007).

(19.) M.N.V.R. Kumar, React. Funct. Polym., 46(1), 1 (2000).

(20.) R.C. Capozza, US Patent 3,989,535 (1976).

(21.) B.M. Min, S.W. Lee, J.N. Lim, Y. You, T.S. Lee, P.H. Kang, and W.H. Park, Polymer, 45, 7137 (2004).

(22.) K. Ohkawa, D. Cha, H. Kim, A. Nishida, and H. Yamamoto, Macromol. Rapid Commun., 25, 1600 (2004).

(23.) X. Geng, O.-H. Kwon, and J. Jang, Biomaterials, 26(27), 5427 (2005).

(24.) P. Sangsanoh and P. Supaphol, Biomacromolecules, 7(10), 2710(2006).

(25.) J.D. Schiffman and C.L. Schauer, Biomacromolecules, 8(9), 2665 (2007).

(26.) M.M. Hohman, M. Shin, G. Rutledge, and M.P. Brenner, Phys. Fluids, 13(8), 2201 (2001).

(27.) M.M. Hohman, M. Shin, G. Rutledge, and M.P. Brenner, Phys. Fluids, 13(8), 2221 (2001).

(28.) A.L. Yarin, S. Koombhongse, and D.H. Reneker, J. Appl. Phys., 89(5), 3018 (2001).

(29.) K. Ogawa, Agric, Biol. Chem., 55(9), 2375 (1991).

(30.) M. Wada and Y. Saito, J. Polym. Sci., Part B: Polym. Phys., 39(1), 168 (2001).

(31.) F. Feng, Y. Liu, and K. Hu, Carbohydr. Res., 339, 2321 (2004).

(32.) J.D. Schiffman and C.L. Schauer, Biomacromolecules, 8(2), 594 (2007).

(33.) G. Cardenas, G. Cabrera, E. Taboada, and P. Miranda, J. Appl. Polym. Sci., 93, 1876 (2004).

(34.) M. Jaworska, K. Sakurai, P. Gaudon, and E. Guibal, Polym. Int., 52(2), 198 (2003).

(35.) B. Focher, P.L. Beltrame, A. Naggi, and G. Torri, Carbohydr. Polym., 12, 405 (1990).

(36.) B. Focher, A. Naggi, G. Torri, A. Cosani, and M. Terboje-vieh, Carhohydr. Polym., 43 (1992).

(37.) P.E.J. Flewitt, R.K. Wild, Physical Methods for Materials Characterization, B. Cantor, Ed., Institute of Physics Publishing, Philadelphia (1994).

(38.) N.L.B.M. Yusof, L.Y. Lim, and E. Khor, Carbohydr. Res., 339(16), 2701 (2004).

(39.) G. Qun, W. Ajun, and Z. Yong, J. Appl. Polym. Sci., 104(4), 2720 (2007).

(40.) R.A.A. Muzzarelli, P. Morganti, G. Morganti, P. Palombo, M. Palombo, G. Biagini, M. Mattioli Belmonte, F. Giantomassi, F. Orlandi, and C. Mazzarelli, Carbohydr. Polym., 70(3), 274 (2007).

(41.) L. Sun, Y. Du, J. Yang, X. Shi, J. Li, X. Wang, and J.F. Kennedy, Carhohydr. Polym., 66(2), 168 (2006).

(42.) J. Sriupayo, P. Supaphol, J. Blackwell, and R. Rujiravanit, Carbohydr. Polym., 62(2), 130 (2005).

(43.) S.K. Choudhari, A.A. Kittur, S.S. Kulkarni, and M.Y. Kariduraganavar, J. Membr. Sci., 302(1-2), 197 (2007).

(44.) K. Kurita, Y. Koyama, and A. Taniguchi, J. Appl. Polym. Sci., 31(5), 1169 (1986).

(45.) C. Tual, E. Espuche, M. Escoubes, and A. Domard, J. Polym. Sci. Part B: Polym. Phys., 38(11), 1521 (2000).

(46.) O.A.C. Monteiro and C. Airoldi, Int. J. Biol. Macromol., 26, 119(1999).

(47.) K.I. Draget, Polymer Gels and Networks, 4(2), 143 (1996).

(48.) J.Z. Knaul, S.M. Hudson, and K.A.M. Creber, J. Polym. Sci. Part B: Polym. Phys., 37(11), 1079 (1999).

(49.) A.F. Groboillot, C.P. Champagne, G.D. Darling, D. Poncelet, and R.J. Neufeld, Biotechnol. Bioeng., 42(10), 1157 (1993).

(50.) E.S. Stevens, in Green Plastics: An Introduction to the New Science of Biodegradable Plastics, Princeton University Press, Princeton, 272 (2002).

(51.) W.L. Teng, E. Khor, T.K. Tan, L.Y. Lim, and S.C. Tan, Carbohydr. Res., 332(3), 305 (2001).

(52.) T.K. Sini, S. Santhosh, and P.T. Mathew, Carbohydr. Polym., 342, 2423 (2007).

(53.) S. Mima, M. Miya, R. Iwamoto, and S. Yoshikawa, J. Appl. Polym. Sci., 28(6), 1909 (1983).

(54.) R. Minke and J. Blackwell, J. Mol. Biol., 120(2), 167 (1978).

(55.) T. Miyazawa, T. Shimanouchi, and S.I. Mizushima, J. Chem. Phys., 24, 408 (1956).

(56.) B. Focher, A. Naggi, G. Torri, A. Cosani, and M. Terbojevich, Carbohydr. Polym., 17(2), 97 (1992).

(57.) Z. Szabo, K. Jost, K. Soos, M. Zarandi, J.T. Kiss, and B. Penke, J. Mol. Struct., 480-481, 481 (1999).

(58.) M. Hasegawa, A. Isogai, F. Onabe, and M. Usuda, J. Appl. Polym. Sci., 45(10), 1857 (1992).

Jessica D. Schiffman, Laura A. Stulga, Caroline L. Schauer

Department of Materials Science and Engineering, Drexel University, Philadelphia, Pennsylvania 19104

Correspondence to: Caroline L. Schauer; e-mail: [e-mail]. edu

Contract grant sponsor: National Science Foundation (NSF)-Integrative Graduate Education and Research Traineeship (NSF IGERT); contract grant number: DGE-0221664; contract grant sponsor: Graduate Assistance in Areas of National Need-Drexel Research and Education in Advanced Materials (GAANN-DREAM); contract grant number: P200A060117; contract grant sponsor: NSF: contract grant number: EEC 0552711.

DOI 10.1002/pen.21434