New insights on solvent implications in flow behavior and interfacial interactions of hydroxypropylmethyl cellulose with cells/bacteria

2.1 Materials

Hydroxypropyl methylcellulose (HPMC) was purchased from the Sigma-Aldrich Company (Darmstadt, Germany). According to product specifications, HPMC has a viscosity of 40–60 cP, 2% in water (20°C). Acetic acid, N,N-dimethylformamide (DMF) and N-methylpyrrolidone (NMP) with high purity (99.8% purity), as well as test liquids [ethylene glycol (EG) and formamidee (FA)] used in contact angle determinations were achieved from the Sigma-Aldrich Company (Darmstadt, Germany).

2.2 Sample preparation and characterization

The sample solution is prepared by dissolving HPMC in different solvents: water, acetic acid, DMF and NMP, resulting a concentration of 15%. The HPMC films, with a thickness of about 50 μm, were prepared by the solution casting on a smooth glass substrate. In the first stage, the samples were dried in a saturated atmosphere of the used solvents. After 1 week, the films were introduced into a vacuum oven for 48 h, at high temperatures to remove the remaining solvent.

The flow properties of HPMC were determined on a Bohlin CS50 rheometer (Malvern Instruments, Worcestershire, UK) equipped with cone-plate geometry (4°/40 mm). Viscosity tests were taken in 0.1–100 s^–1 shear rates interval (25°C). Oscillatory tests were performed in the linear viscoelastic region by selecting a shear stress of 2 Pa and then frequency sweep experiments were made in the range of 0.01–15 Hz.

Contact angles with EG and FA were determined on a CAM-101 contact angle device (KSV Instruments Ltd, Helsinki, Finland). Liquid drops of about 1 μl were cast, with a Hamilton syringe, on the analyzed sample surface. In order to establish the balance of the forces involved, the contact angle reading was obtained 20 s after deposition of the drop on the surface of the samples. For each liquid drop, 10 photos were recorded at an interval of 0.016 s and each angle value corresponds to an average of five measurements. For each test liquid, different surface zones were chosen to achieve a statistical result. All the contact angle measurements were performed at room temperature.

3.1 Rheological analysis

Rheology is a sensitive tool to the changes occurring in the polymer microstructure as a result of the interactions taking place in the system. The solvents considered in this study have distinct characteristics in terms of quality, size, polarity and hydrogen bond forming ability. The basic features of water, acetic acid, DMF and NMP are taken from literature (https://www.accudynetest.com/solubility_table.html) and are presented in Table 1.

In order to assess the solvent quality, we have determined the solubility parameter (δ) of HPMC by using the formalism suggested by Bicerano in his work (24). The approach for estimation of polymer solubility parameter relies on determination of atomic and valence connectivity indices. These parameters are further used in establishing the values of cohesive energy (E_coh) and molar volume at room temperature (V). The procedure concerning the calculation of connectivity indices and structural increments affecting the values of E_coh and V, was previously detailed (24). The equations used in our investigation were adapted to the analyzed polymer structure, as seen in relations [1–3]:

where ⁰χ, ⁰χ^ν, ¹χ, ¹χ^ν represent the zeroth- and first-order connectivity indices, N_OH is the number of hydroxyl groups, N_ether denotes the number of ethers in the structural unit, N_cyc describes the number of nonaromatic rings that lack double bonds on their edges.

From our calculations it was shown that: ⁰χ=18.40, ⁰χ^ν=18.81, ¹χ=15.43, ¹χ^ν=10.88, N_OH=4, N_ether=4 and N_cyc=2. For HPMC a cohesive energy of 2.314·10⁵ J/mol, a molar volume of 365.013 ml/mol and a solubility parameter of 25.18 were obtained. Analyzing the solubility parameters of both polymer and each solvent used, it can be seen that solvent quality ranges in the following order: DMF>NMP>acetic acid>water. It is well known that in a good solvent the macromolecular coil is less compact determining a different chain conformation in comparison with a poor solvent (2). These aspects are reflected in the resistance of a polymer solution to flow under shearing conditions. The viscosity (η) dependence on the shear rate, for all examined HPMC samples, is depicted in Figure 1. It can be observed that the solvent polarity influences the magnitude of viscosity at shear rate values close to zero. Thus, at shear rate of 0.1 s⁻¹ the viscosity increases in the following order: NMP<acid acetic<DMF<water. Even if acetic acid is less polar than DMF and NMP, its protic character favors besides polar interactions, the formation of hydrogen bonds, which determine higher viscosity. Literature (25), (26) reports that such interactions determine the increase of solution viscosity.

Figure 1:

The viscosity (red curve) and shear stress (blue curve) dependence on shear rate for the HPMC solutions in acetic acid, water, NMP and DMF.

The shape of flow curves is distinct depending on the solvent used. One may remark in Figure 1 that solvents with higher δ_p, like water and acetic acid, exhibit a flow behavior consisting in a Newtonian plateau, followed by shear thinning domain. Moreover, due to the fact that water molecules are smaller and more polar, they can interact more easily with HPMC, leading to a longer Newtonian region than that recorded for acetic acid. Conversely, for the other two solutions no constant viscosity domain is noticed. The polymer solutions in DMF and NMP present only shear thinning behavior, regardless the applied shear rate. The result can be explained by considering that water and acetic acid are protic polar solvents that might give their hydrogen more easily to HPMC to form hydrogen bonds. This aspect is not possible for the aprotic and dipolar solvents, namely DMF and NMP. Also, the higher polarity of DMF generates bigger polymer/solvent interactions than polymer/polymer ones, which facilitate the chain disentanglement in the case of HPMC/DMF solution as seen in the higher slope of viscosity. Given the fact that HPMC/water interactions are assumed to be stronger than those from the HPMC/acetic acid system, one can explain the longer Newtonian zone as shearing disrupts harder to a great extent the hydrogen bonds from the sample. The shear stress (σ), which is defined as the deforming force per unit area, ranges with the shear rate, as depicted in Figure 1. For water and acetic acid solvents, the shear stress range has a slope close to unity in the constant viscosity domain, and begins to decrease once the thinning behavior appears. In the case of the DMF and NMP solvents, the thinning shear flow affects the linearity of the shear stress curve and reduces its slope.

The close analysis of viscosity curves provides information on the solution processability into thin films as it determines its thickness and uniformity (27). These aspects are known to influence the morphology and physico-chemical features of polymer films. When preparing polymer films from solutions by the casting method, the thickness profile is affected by the viscosity behavior under deformation as it is less probable to obtain uniform dried films from non Newtonian fluids (27). In this context, it can be interpreted that HPMC samples that exhibit Newtonian behavior are more easily processed into solid films with a uniform thickness profile, being more suited for bio-applications.

The interactions occurring in polymer solutions are also described by flow activation energy (E_a). This parameter is obtained from the dependence of viscosity (at shear rates close to zero) on temperature (T). The application of the Arrhenius equation to ln η vs. 1000/T dependence, one may extract the activation energy of flow processes from the slope. Figure 2 shows the influence of solvent nature on the viscosity dependence on temperature. The magnitude of E_a values gives useful insights concerning the energetic barrier required so the flow process can take place. For the studied solutions, it can be seen in Figure 2 that ln η vs. 1000/T curve presents different slope values as a result of the solvent nature. In other words, HPMC chains exhibit different mobility as a function of the polar and hydrogen bonding interactions from the system. Flow activation energy is higher for HPMC in solvents with higher polarity (water and DMF), except for acetic acid. The latter is due to its protic character and smaller molecular size that favor higher interactions that raise flow energy barriers.

Figure 2:

The dependence of viscosity (taken at 0.1 s⁻¹) versus reciprocal temperature for HPMC solutions in different solvents.

The viscoelastic properties of the investigated solutions are examined through shear oscillatory tests. The magnitude of storage modulus (G′) in regard with the loss modulus (G″) reveals if elastic or viscous character prevail. For all analyzed samples, it can be remarked that at low frequencies the loss modulus is higher than the storage one, while at a point at which G′ become equal to G″, the elastic character of the samples is predominant. The shear oscillatory tests recorded for HPMC solutions are depicted in Figure 3.

Figure 3:

The dependence of rheological moduli on shear frequency for the studied HPMC solutions.

The viscoelastic properties were measured in the linear range by performing the controlled-stress test, so the samples are deformed in an oscillatory shearing regime until they reach a maximum strain, where G′ and G″ are independent of the imposed stress. The frequency sweep measurements are illustrated in Figure 3, showing that all HPMC solutions are characterized by a transition from the viscous flow to the elastic flow. The rheological moduli dependence on shear frequency (f) is described by a power law, G′, G″~f^x (28). The x exponent may be different or equal for both moduli, depending on the microstructure changes determined by interactions from the system. For typical viscoelastic fluids, one can observe that G″~f², while G″~f¹ (28). In the case of our samples, the power law exponent is changing as a result of the solvent’s nature. The elastic modulus (G′) is related to the strain energy reversibility contained in the material, and is affected by the amount and intensity of the interactions in the system. Indeed, for all polymer solutions, it was noticed that G′ become more sensitive to the HPMC/solvent interactions. For HPMC solutions in water and acetic acid, the slopes of G′ vs. shear frequency are slightly smaller (1.02 and 1.11) in comparison with those of HPMC solutions in DMF and NMP (1.18 and 1.21). The result is supported by literature data (26), which indicate that the presence of hydrogen bonds determine the lowering of power low indices from 2 and 1 to almost 0.5. In addition, it was reported that such interactions generate a decrease of the values corresponding to the frequency at which rheological moduli become equal (26). This is also valid for our studied samples as indicated in Figure 3. In the high shear frequency region, the elastic modulus overcomes the viscous one, delimiting a transition towards solid-like fluids. The crossover frequency (f_c) of shear moduli is distinct as a function of solvent type. The overlap point is lower when preparing HPMC solutions in polar and protic solvents, whereas it increases for solutions obtained in aprotic and dipolar solvents. Thus, the values of f_c are presented in Figure 3 and are ranged in the following order: water<acetic acid<DMF<NMP. The overlap frequency contains information on the relaxation time, which in turn reveals the extent to which the polymer structure is modified in response to shear flow. After cessation of shear forces, the polymer solution begins to relax and again takes its initial form. A high value of relaxation time and implicitly a low value of f_c means that the sample requires a long time for recovery because its rigid component is bigger than the elastic one. The solutions prepared in solvents that facilitate more pronounced polar and hydrogen bond interactions with HPMC are characterized by a higher relaxation time. Another aspect noticed in Figure 3 is referring to the magnitude of rheological moduli at high shear frequencies above the overlap point. Water and acetic acid solvents lead to higher values of G′ above 10³, as a result of the more intense hydrogen bonding interactions.

3.2 Surface wettability

The surface tension values of the HPMC films, prepared in distinct solvents, were determined by using the equations [4] and [5] developed by Fowkes (29):

where subscripts “l” and “s” denote the liquid–vapour and the solid-vapour interface, while superscripts “d” and “p” refer to the dispersed and the polar components of the surface tension.

The surface tension components of the test liquids were extracted from the literature (30) and used in our calculations. The experimental values of the EG and FA contact angles with HPMC samples are listed in Table 2. It can be noticed that surface tension values of solid specimens are influenced by the type of solvent involved in film processing. Particularly, the polar component, γsp, is ranging from 9.71 dyn/cm for HPMC/acetic acid film to 20.49 dyn/cm for HPMC/water.

Analyzing the results from Table 2, one may observe that the polar surface tension is higher than the dispersed one. This is in agreement with the work of Kraisit and collaborators (31), which reported a prevalent polar character of HPMC film’s surface. In addition, the polar surface tension is varying in the same order as the solvent polarity. The idea is supported by the work of Wang and co-workers (32), which showed that the polymer surface wettability can be affected by the solvent nature. In other words, during the transformation of polymer solutions in polar solvents to solid films, the solvent evaporation leaves at the polymer surface the polar groups that interacted with the solvent through hydrogen bonding. Based on these aspects, one may assume that the polar groups present in the HPMC structure (OH or -CH₂-CHOH-CH₃) are able to interact with the solvents, like water and DMF, through polar forces and intermolecular hydrogen bonds and, after the removal of the solvent, these groups are distributed mainly at the polymer film surface. Conversely, the solvents with reduced polarity, like NMP and acetic acid, determine less intermolecular interactions and thus HPMC chains are able to shrink to form intramolecular interactions, leaving the hydrophobic groups at the sample surface after solvent evaporation.

3.3 Spreading/adhesion interactions of HPMC with cells or bacteria

Surface energy parameters are very important in biomedicine as they allow the estimation of the spreading ability of biological materials on polymer surfaces. For this purpose, we have used equation [6] to achieve the values of the work of spreading (Ws) of different bio-agents on the HPMC film’s surface. According to equation [6], Ws represents the difference between the work of adhesion (Wa) and the work of cohesion (Wc), as follows:

where the index “b” indicates the biological materials considered to interact with the HPMC films, here under study.

Depending on the bio-application being pursued, one should be focused on obtaining a specific balance between cohesion and adhesion interactions at the bio-interface. For cell culture purposes, it is desirable to achieve a high adhesion with the polymer substrate in order to ensure the growth and differentiation of cells in proper conditions, as required by tissue regeneration. On the other hand, when polymers are involved in implants or biomedical instrumentation manufacturing, the adhesion with some dangerous microorganisms is unfavorable for human health, so it must be avoided. For such applications, one should pursue enhancing the cohesion interactions with unwanted microorganisms. We have investigated the work of the spreading of Escherichia coli, Staphylococcus epidermidis and fibroblasts on the studied HPMC samples. Figure 4 shows the results concerning their spreading behavior by highlighting the effect of the used solvents in processing stage of HPMC.

Figure 4:

The work of spreading of (A) E. coli, (B) S. epidermidis and (C) fibroblasts on the studied HPMC samples as a function of the used solvents in the processing stage.

The work of the spreading of E. coli and S. epidermidis on the HPMC specimens has negative values, meaning that the work of adhesion is smaller than that of cohesion. The samples prepared in acetic acid and NMP lead to the highest cohesion owing to their lower surface polarity, so they attract less these two microorganisms comparatively with films obtained in water and DMF. Furthermore, when comparing the absolute values of Ws, one may notice that the level of cohesion of E. coli on cellulosic polymer surface is higher than that of S. epidermidis (Figure 4A and B). The work of spreading is smaller when assessing the interactions at HPMC/fibroblasts interface. For a better understanding, we have also included in Figure 4C the work of adhesion at the polymer/cell interface. This parameter ranges for HPMC samples in the following order: water>DMF>acetic acid>NMP. It can be observed that processing HPMC in polar solvents, such as water or DMF, leads to higher values of the work of adhesion with fibroblasts. These samples are more adequate as cell growth supports in tissue engineering. One may conclude that processing the cellulose derivative in less polar solvents facilitates the repulsion of unwanted microorganisms, while strong polar solvents are suitable for increasing adhesion forces with cells. Thus, surface wettability can be adapted through the solvent features to achieve the desired spreading behavior at the bio-interface.