Printer Friendly

Alginate gelling process: use of bivalent ions rich microspheres.

INTRODUCTION

Natural polymers are emerging as new smart materials, in particular the use of alginate, a natural polymer extracted from brown algae, has acquired great relevance due to its low cost, biocompatibility, biodegradability, and easiness of functionalization with many different materials and compounds [1]. This polymer has shown interesting characteristics in many fields and applications, from pharmaceutical to environmental protection [2, 3]. This wide range of applications is due to its ability to form a physical hydrogel, when bivalent cations ([Me.sup.2+]) are added to the alginate solution [4]. The polymer chain is made of (l,4)-linked [beta]-D-mannuronic acid (M). and its C-5 epimere, a-lguluronic acid (G), able to connect to each other in long sequences of M-Blocks, G-Blocks, and MG-Blocks [5, 6]. The gelling capacity is directly correlated to the amount of G-Blocks present along the chain, because only the [alpha]-L-guluronic acid has an active role in the formation of crosslinks in the hydrogel. The G-Blocks of the polymer chain form cavities, which work as a binding site for the bivalent cations and then arrange themselves all around creating a structure comparable to an "egg box" [7].

The gelling process of sodium alginate (SA) is usually achieved using bivalent cations in two different ways: the diffusion method and the internal gelation method [8]. In the first case, bivalent ions are introduced from an outer reservoir (i.e., saturated solution); the gelation happens almost instantaneously and it creates an inhomogeneous distribution of the crosslinks inside the alginate gel. In the second case bivalent ions are added in an inactive form to the alginate solution (i.e., chelated into EDTA or citrate) [9]. By lowering the pH, the ions are slowly released and a homogeneous network is created.

An innovative approach to develop a simple and practical method to release bivalent ions in a controlled way, without adding any additional compound, was presented [10]. The method consisted in the use of SA microbeads, with controlled shape and dimensions, enriched with different amounts of the [Ca.sup.2+] ions, obtained by electrospraying. The microspheres were then dispersed into the alginate solution, with the double function of gelling germs and ions source. This method allowed a gelation with the same material that had to be gelled, as crosslinking agent.

As reported in literature, alginate's affinity toward the different divalent ions decreases as follows: [Pb.sup.2+] > [Cu.sup.2+] > [Cd.sup.2+] > [Ba.sup.2+] > [Sr.sup.2+] > [Ca.sup.2+] > [Co.sup.2+], [Ni.sup.2+], [Zn.sup.2+] > [Mn.sup.2+] [11-13].

In this article, the results obtained with the new methodology using [Ca.sup.2+] [10] as a crosslinker, are compared with those obtained using two different bivalent ions: Ba~" and [Mg.sup.2+] [14, 15].

According to the available data in literature, the gelling process, for the [Mg.sup.2+]/alginate system, does not occur [16, 17] due to the weak polymer-ion interactions [18, 19]; in this study, a stable [Mg.sup.2+]/alginate gel was achieved using a solvent/nonsolvcnt system to collect the microspheres.

The morphology and the ions content of the microspheres were evaluated. The gels were investigated with rheological measurements to show the effect of the different ions added on the final dynamic-mechanical proprieties, in particular the change in the storage modulus (G') and loss modulus (G").

MATERIALS AND METHODS

Materials

SA, medium viscosity, from brown algae, with a molecular weight ranging from 80,000 to 120,000 g/mol and composed of ~61% mannuronic acid and 39% guluronic acid (M/G ratio of 1.56) was purchased by Sigma-Aldrich and used without any further purification. Anhydrous granular calcium chloride Ca[Cl.sub.2] ([greater than or equal to] 93.0%), barium chloride Ba[Cl.sub.2] x [H.sub.2]O ([greater than or equal to] 99%), magnesium sulfate MgS[O.sub.4] ([greater than or equal to] 99%), and absolute ethanol [greater than or equal to] 99.8% were purchased from Sigma-Aldrich. Hydrochloric acid was purchased from Sigma-Aldrich ACS reagent (37%).

Solutions Preparation

SA solutions were prepared at 2%-3% w/w in deionized water, with gentle stirring, until complete dissolution of the SA powder; the solutions were then put into tubes and centrifuged at about 5,000 rpm for a maximum of 5 min to eliminate the air trapped inside. The solutions were then stored in the fridge at 4[degrees]C to avoid the biodegradation of the alginate.

For the salt solutions, different concentrations were prepared. Calcium chloride and barium chloride solutions were prepared in deionized water, with gentle stirring, until complete dissolution, at 740 and 42 g/l in water at 20[degrees]C, respectively. Magnesium sulfate solutions were prepared at 10 g/l in 60:40 water/ ethanol at 20[degrees]C. The different solutions were selected on the basis of the affinity of alginate for ions: [Ba.sup.2+] > [Ca.sup.2+] [??] [Mg.sup.2+].

Microspheres and Gel Preparation

An electrospray system was used to create microspheres with a suitable diameter to be well dispersed inside the SA solution (2% w/w) [10, 20, 21]. The system, purposely developed for this aim and sketched in Fig. 1 was made up by a Harvard apparatus syringe volumetric pump that provides a controlled flow on a 5 ml glass syringe of the FORTUNA OPTIMA series, purchased by Sigma-Aldrich. with an internal diameter of 11.60 mm. The SA was pumped inside a Teflon pipe, which had a filter and a needle tip on the extremity. SA solution was sprayed into a solution of bivalent ions contained in a becker, placed over the anode, at the distance of 5 cm to the needle. The SA drops became gels and were enriched with bivalent ions over the time. The parameters (electric field applied between the needle tip and the anode, and the flow applied to the solution by a volumetric pump) used for the formation of the different microspheres were summarized in Table 1. To obtain the enriched bivalent ions microspheres the collecting solutions contained [Ba.sup.2+], [Ca.sup.2+], or [Mg.sup.2+].

Barium ions, caused by their higher affinity toward alginate, form stronger gels compared with the more used calcium ions. Therefore to obtain microspheres comparable to those obtained with calcium in terms of shell thickness, that is, able to diffuse bivalent ions through SA solutions, the concentration of the Ba[Cl.sub.2] in the collecting solution was lower than that of Ca[Cl.sub.2].

On the contrary, magnesium sulfate was used as a saturated solution, to collect electrosprayed microspheres, due to the poor gelling strength of [Mg.sup.2+] ions [13, 14]. To avoid microspheres dissolution, inside the aqueous environment, a 60:40 water/ethanol mixture was used, where ethanol is a nonsolvent for the alginate [22]. The 60:40 ratio is the appropriate compromise as it allows to solubilize enough magnesium salt and not solubilize the alginate microspheres.

The microspheres were maintained in solution from 1 to 7 days to achieve the highest ions content inside, were then filtered and subsequently washed with absolute ethanol, to eliminate the aqueous solution, which causes the formation of aggregates. Later, the microspheres were put in an oven at around 40[degrees]C to evaporate the ethanol and finally stored, at room temperature, in a dryer to avoid the absorption of humidity.

The alginate gels were prepared starting from an SA solution at 3% w/w adding different amounts of ions rich microspheres and slowly mechanically stirred for about 5 min [10]. After that, the solutions were kept at 4[degrees]C until the gelling process was completed. The gelling time varied depending on the ions used, from a minimum of 20 min for the Ba_gels to up to several hours for the Mg_gels.

Characterization

A polarized Reicther Polyvar Pol microscope equipped with a 40X objective lens was used for the evaluation of shape and size of the different microspheres. Micrographs were acquired with a computer-controlled digital camera (Motic). For the evaluation of the average diameter of the different types of microspheres, a statistical approach was chosen. No less than 400 microbeads were photographed under the polarized optical microscope and the diameters were measured with an open source software imageJ.

To quantify the amount of divalent ions present inside the microspheres, a sample of about 10 mg of dried microspheres was put in the oven at 650[degrees]C for 8 h to complete the degradation of the alginate gel. The resulting powder was then collected with 1 ml of hydrochloric acid, brought to 10 ml volume with Milli-Q water, and then suitably diluted with Milli-Q water. Finally, the solutions were analyzed using Atomic absorption spectroscopy (AAS) performed with Variant Spectra AA55B, with external calibration. An SA powder sample was used as a reference.

Rheological measurements were performed using an Anton Paar Physica MCR 301 Rheomcter (Anton Paar, GmbH. Germany) equipped with 25 mm parallel plate geometry (PP25), with 2 mm gaps. A Peltier heating system for the accurate control of the temperature was used, set at 25.00[degrees]C [+ or -] 0.01[degrees]C for all measurements. In order to ensure that all measurements were performed within the linear viscoelastic region, amplitude sweep tests (AS), with deformation range ([gamma]) from 0.01 up to 100% at fixed frequency of 1 Hz, were performed. Information about the storage modulus (G') and the loss modulus (G") as a function of the frequency, which was varied from 0.1 to 100 Hz at a fixed deformation ([gamma]) of 0.1% were obtained with frequency sweep tests (FS). A time sweep test, in order to evaluate the gelling point, was performed at fixed frequency of 0.1 Hz and fixed deformation ([gamma]) of 0.5%, with a time range from 0 to 20 min.

RESULTS AND DISCUSSION

Evaluation of Dimension and Ions Content in the Microspheres

The electrospray parameters were set, on the basis of previous work [10], to obtain microbeads with spherical shape and controlled dimensions (Fig. 2a-c). More in detail, the flow and the electric field affected the dimensions of the microbeads: higher flow generated higher diameters; instead higher field led to lower diameters. In Table 1 are reported the optimized parameters for each cation.

The evaluation of the microbeads average diameter was done using a statistical approach. The recorded values of the diameters were processed using two different statistical tests: Median Absolute Deviation tests [23] to verify the presence of outliers from collected data and the Kolmogorov-Smirnov [24-26] tests to verify the theoretical continuous distribution of the populalion. After these preliminary evaluations about the quality of the collected data, a Gaussian distribution of the diameter of the samples was obtained, directly related to the experimental data (Fig. 2d-f).

The selected diameters of the microspheres, chosen for the three bivalent ions (Table 1) were related to two main aspects: their affinity with the alginate and the ratio between crosslinking agent (i.e., bivalent ions) and polymer (i.e., alginate) required. A very strong gelling agent (i.e., [Ba.sup.2+]), which allowed instantaneous gel formation, needed a bigger radius to reduce the exchange specific surface between the microspheres and the SA solution in order to slow down the diffusion process. In this way it was possible to distribute fewer microspheres inside the solution and obtain a stable gel with homogeneous crosslinking, avoiding the formation of highly cured areas inside the gel. A medium gelling agent (i.e., [Ca.sup.2+]) needed a smaller radius to be homogeneously distributed inside SA solution. The lower volume of the microspheres makes it necessary to have a higher number of diffusion points inside the SA solution. A weak gelling agent (i.e., [Mg.sup.2+]), as well as medium gelling agent, needed a small radius to be homogeneously distributed in SA solution [13]. The very low affinity of the [Mg.sup.2+] toward SA is the reason why a very high number of diffusion points was used.

The ions content inside the microspheres was evaluated by carrying out an AAS analysis. The results are shown in Table 2. The microspheres showed a tendency to absorb a higher quantity of [Ca.sup.2+] compared with [Ba.sup.2+], this was confirmed by the ion affinity with the polymer [8, 9]; in fact [Ba.sup.2+] ions instantaneously created a high crosslinkcd shell around the microspheres, that made difficult the further diffusion inside the spheres. After 2 days, microspheres contained around the 28% w/w of [Ba.sup.2+] ions, at the adsorption plateau. However, [Ca.sup.2+] ions, due to the lower ion/polymer affinity, did not create this highly crosslinked shell and the permanence in a saturated solution allowed a higher absorption, till a plateau value of 43% w/w. Magnesium confirmed its low ion/polymer affinity and the absorbed quantity was much lower compared with the other two ions. Nevertheless, it was possible to observe a gelling action in the water/ethanol environment and a plateau value was reached at around 8% w/w.

Alginate Gels

To evaluate the effectiveness of the gelling process, five gels crosslinked by different [Me.sup.2+] with concentrations of [Me.sup.2+]/alginate solution ranging from 0.4 to 0.6% w/w were characterized (Table 3). The third column of the table shows the number of microspheres used as diffusion points per SA solution weight. It is evident that these numbers increased from [Ba.sup.2+] to [Mg.sup.2+] in parallel with the decrease of the affinity. Moreover, even adding different microsphere numbers, gels with a similar concentration of [Me.sup.2+]/alginate solution were obtained, as a result of the different amount of ions inside the microspheres. More in details, with and [Ca.sup.2+] the characterized gels had two different concentrations: a low (Me_gel_1) and a high (Me_gel_h) ions content. The lower affinity of the [Mg.sup.2+] with alginate did not allow to achieve concentrations >0.40% w/w (Mg_gel_1).

Figure 3 shows the good transparency of the gels. Only the Mg_gel_1 has a macroscopically mat appearance due to the high quantity of microspheres needed to obtain a stable gel.

Rlieological Tests

To follow the crosslinking process of [Me.sup.2+]/alginate mixtures, a time sweep test was carried out at a fixed frequency and deformation. Typical gelation profiles for alginate solutions in the presence of bivalent ions are shown in Figs. 4 and 5. The different behaviour of [Mg.sup.2+] in comparison with the other two ions, in the gels formation was evident. In Fig. 4, initially, the alginate solution showed G" > G' with values of 100 and 80 Pa, respectively.

Both moduli gradually increased, as a function of time, suggesting the slow occurrence of interactions between the alginate chains. After about 15 min. G" (140 Pa) was surpassed by G' (145 Pa), indicating that the gelation point was achieved. In the magnification of Fig. 4 the semilogarithmic plot was chosen to clarify the crossover, at about 10 min. at 136 Pa.

When microspheres of [Ca.sup.2+] and [Ba.sup.2+] were used at similar concentrations (0.40-0.50% w/w), the gel point was not detectable by the rlieological measurements, because of the faster crosslinking process [27]; in Fig. 5a G' > G" was shown in all the investigated time range. The different affinities of divalent ions were reflected in the mechanical properties of the gels. In Figs. 4 and 5a. it was observed that the Ba_gel modules were higher (about [10.sup.4] Pa) than those of the Ca_gel (about [10.sup.3] Pa) and that the Mg_gel modules were much lower (about 10" Pa) than the previous ones.

In the case of the calcium ion (Fig. 5b), the only chance of observing the gel point was to use a very low ions concentration (0.02-0.04% w/w). Using these ratios the crossover was observed in a very short time, about 1 min, but the obtained gel showed poor mechanical properties.

However, it was not possible to observe any gel point using the barium ion.

The FS test, carried out at complete gelification, of alginate gels with high (Me_gel_h) and low (Me_gel_1) ions content is shown in Fig. 6a and b, respectively.

Due to the lower affinity of the [Mg.sup.2+] with alginate, and therefore the impossibility to obtain microspheres rich in Mg2 + >8% w/w, a gel with high magnesium content was not obtained (Fig. 6a).

After 2 h from the addition of [Mg.sup.2+] rich microspheres to the SA solution, a stable gel was obtained, having a G' comparable to the one of Ca_gel_1 and slightly lower than that of Ba_gel_1, both obtained within a few minutes with the same low ions content (Fig. 6b).

It is important to state that when the upper measurement plate approaches the gel deposited onto the lower plate, the material is partially squeezed out from the head-plate gap and a minimum amount of water is released from the gel network. Therefore, during the measurements, a small increase of G' modulus could be observed. This phenomenon was due to the water loss caused by both the evaporation and the little pressure applied over the gel. In Fig. 6b, this behavior is more evident for the weakest Mg gel where the water can easily go out. Nevertheless, to make the measurements reliable, we always used samples with the same swelling degree on the plate, and we always maintained the same measurement conditions.

Gel consistency was maintained also after the tests, highlighting the general stability of these systems.

CONCLUSIONS

This paper describes a new methodology for obtaining a controlled gelling process of SA using alkaline earth metal ions, including a typical nongelling ion: [Mg.sup.2+].

The use of alginate microspheres, obtained by electrospray and enriched with different bivalent ions ([Ca.sup.2+], [Ba.sup.2+]. and [Mg.sup.2+]) was described; in particular, it was possible to prepare Mg_ microspheres. as a result of the water/ethanol collecting solution.

The presence of ethanol. a nonsolvent for alginate, was helpful to solubilize enough magnesium salt and to avoid microspheres dissolution, allowing the slow crosslinking by [Mg.sup.2+] ions.

The microspheres were used as ions diffusion points into the SA solution crosslinking process. To control the gel formation, different microsphere numbers were added to the SA solution, taking into account the different amounts of ions inside the microspheres. With this method, gels with similar concentration of the [Me.sup.2+]/alginate solution were obtained also with magnesium ions, which have a very low affinity with the alginate.

Alginate gels were characterized, from a rheological point of view and the Mg_gel showed a G' comparable to the one of Ca_gel and slightly lower than that of Ba_gel.

The mechanical behavior of the gel is an indication of the lifetime of the material, and the choice of the alginate/cation pair influences the gelling time offering the opportunity to use the hydrogel in different application fields.

ACKNOWLEDGMENT

The authors would like to thank Erasmus+ Project for the support and A. Dodero for the cooperation during the experiments.

REFERENCES

[1.] J.S. Yang, Y.J. Xie, and W. He, Carbohydr. Polym., 84. 33 (2011).

[2.] S. Cataldo, G. Cavallaro, A. Gianguzza, G. Lazzara. A. Pettignano, D. Piazzese, and I. Villaescusa, J. Environ. Chem. Eng., 1, 1252 (2013).

[3.] K.Y. Lee, and D.J. Mooney, Prog. Polym. Sci., 37. 106 (2012).

[4.] M.B. Stewart, S.R. Gray, T. Vasiljevic, and J.D. Orbell. Carbohydr. Polym., 102. 246 (2014).

[5.] M.B. Stewart, S.R. Gray, T. Vasiljevic, and J.D. Orbell, Carbohydr. Polym., 112, 486 (2014).

[6.] I. Donati, S. Holtan, Y.A. Morch, M. Borgogna, M. Dentini. and G. Skjak-Braek, Biomacromolecules, 6, 1031 (2005).

[7.] P. Sikorski, F. Mo, G. Skjak-Braek, and B.T. Stokke. Biomacromolecules, 8. 2098 (2007).

[8.] K.I. Draget, G. Skjak-Braek, and O. Smidsrod. Int. J. Biol. MacromoL, 21, 47 (1997).

[9.] X. Liu, L. Qian, T. Shu, and Z. Tong, Polymer, 44, 407 (2003).

[10.] S. Vicini, M. Castellano, M. Mauri, and E. Marsano, Carbohydr. Polym., 134, 767 (2015).

[11.] A. Haug, and O. Smidsrod, Acta Chem. Stand., 24. 843 (1970).

[12.] A. Haug, Acta Chem. Scand.. 15, 1794 (1961).

[13.] P. Agulhon, M. Robitzer, J.P. Habas, and F. Quignard, Carbohydr. Polym., 112, 525 (2014).

[14.] F. Topuz, A. Henke, W. Richtering, and J. Groll, Soft Matter, 8, 4877 (2012).

[15.] Y.A. Morch, I. Donati, B.L. Strand, and G. Skjak-Br;ek. Biomacromolecules, 7, 1471 (2006).

[16.] G.R. Seely, and R.L. Hart, Macromolecules, 7. 706 (1974).

[17.] V. Florian-Algarin, and A. Acevedo, J. Pharm. Innov.. 5. 37 (2010).

[18.] I. Donati, A. Cesaro, and S. Paoletti, Biomacromolecules, 7. 281 (2006).

[19.] I. Donati, F. Asaro, and S. Paoletti, J. Phys. Chem. B. 113, 12877 (2009).

[20.] H. Moghadam, M. Samimi, A. Samimi, and M. Khorram. Particuology, 6. 271 (2008).

[21.] P. Shi, P. He, T.K.H. Teh, Y.S. Morsi, and J.C.H. Goh. Powder Technol., 210. 60 (2011).

[22.] J. Li, J. He. Y. Huang, D. Li. and X. Chen, Carbohydr. Polym.. 123, 208 (2015).

[23.] T. Pham-Gia, and T.L. Hung, Math. Comput. Model. 34. 921 (2001).

[24.] Z. Drezner, and O. Turel, Comput. Ind. Eng., 61. 1240 (2011).

[25.] H.L. Koul, and P.K. Sen, Stat. Prob. Lett., 3, 111 (1985).

[26.] E. Reschenhofer, Comput. Stat. Data Anal., 24. 433 (1997).

[27.] B.T. Stokke, K.I. Draget, O. Smidsrod, Y. Yuguchi, H. Urakawa, and K. Kajiwara, Macromolecules, 33, 1853 (2000).

Silvia Vicini, (1) Marco Mauri, (1) Joanna Wichert, (2) Maila Castellano (1)

(1) Department of Chemistry and Industrial Chemistry, University of Genova, Genova 16146, Italy

(2) Wroclaw University of Technology, Wroclaw, 50-370, Poland

Correspondence to: Maila Castellano; e-mail: maila.eastellano@unige.it

DOI 10.1002/pen.24552

Caption: FIG. 1. Schematic representation of the electrospray system.

Caption: FIG. 2. Optical microscopy images (40X) of the microbeads obtained with electrospray methodology: (a) Ca_microspheres; (b) Ba_microspheres; and (c) Mg_microspheres. Correspondent statistical evaluation of the diameters is reported below in (d-f). respectively. IColor figure can be viewed al wileyonlinelibrary.com]

Caption: FIG. 3. Gels erosslinked by the three types of [Me.sup.2+] microspheres. [Color figure can be viewed at wileyonlinelibrary.com]

Caption: FIG. 4. Storage modulus (G') and loss modulus (G") in the gelation of alginate with [Mg.sup.2+]. In the inset was reported the crossover of the moduli corresponding to the gel point.

Caption: FIG. 5. Storage modulus (G') and loss modulus (G") in the gelation of alginate with [Ca.sup.2+] and [Ba.sup.2+]: (a) in concentration [Me.sup.2+]/[g.sub.sol] = 0.40-0.50% w/w; (b) in concentration [Ca.sup.2+]/[g.sub.sol] = 0.02-0.04% w/w.

Caption: FIG. 6. Storage modulus (G') of alginate gels as a functions of frequency after complete gelification: (a) gels with high ions concentration [M.sup.e2+]/ [g.sub.sol] = 0.50% w/w; (b) gels with low ions concentration [Me.sup.2+]/[g.sub.sol] = 0.40% w/w.
TABLE 1. Electrospray parameters used for the preparation of the
microspheres.

Ions          Voltage (kV)   Flow (ml/min)   Diameter ([micro]m)
[Ba.sup.2+]   5              0.1             460 [+ or -] 20
[Ca.sup.2+]   20             0.01            170 [+ or -] 20
[Mg.sup.2+]   20             0.01            220 [+ or -] 20

TABLE 2. Data of the AAS measurements.

                  Time in the        Concentration
Ions          collecting solution       (% w/w)

[Ba.sup.2+]          1 day          22 [+ or -] 2%
                    2 days          28 [+ or -] 2%
                    3 days          28 [+ or -] 2%
                    4 days          28 [+ or -] 2%
[Ca.sup.2+]          1 day          15 [+ or -] 2%
                    2 days          43 [+ or -] 2%
                    3 days          43 [+ or -] 2%
                    4 days          43 [+ or -] 2%
[Mg.sup.2+]          1 day           5 [+ or -] 2%
                    2 days           6 [+ or -] 2%
                    3 days           6 [+ or -] 2%
                    4 days           7 [+ or -] 2%
                    5 days           8 [+ or -] 2%
                    7 days           8 [+ or -] 2%

Samples of 10 ml were diluted during the analytical procedure.

TABLE 3. Samples for rheological tests.

           Microsphere   [n.sup.[degrees]f of      [Me.sup.2+]/
Sample     (MS) type        MS/[g.sub.sol]       [g.sub.sol] % w/w

Ba_gel_l   MS_460_2d              180                  0.41
Ba_gel_h                          255                  0.58
Ca_gel_l   MS_170_2d             2200                  0.38
Ca_gel_h                         2900                  0.51
Mg_gel_l   MS_220_5d             5600                  0.40
COPYRIGHT 2017 Society of Plastics Engineers, Inc.
No portion of this article can be reproduced without the express written permission from the copyright holder.
Copyright 2017 Gale, Cengage Learning. All rights reserved.

Article Details
Printer friendly Cite/link Email Feedback
Author:Vicini, Silvia; Mauri, Marco; Wicher, Joanna; Castellano, Maila
Publication:Polymer Engineering and Science
Article Type:Report
Date:Jun 1, 2017
Words:4098
Previous Article:Reinforcement effect of cellulose nanocrystals in thermoplastic polyurethane matrices characterized by different soft/hard segment ratio.
Next Article:Physicochemical and antimicrobial properties of silver-doped hydroxyapatite collagen biocomposite.
Topics:

Terms of use | Privacy policy | Copyright © 2018 Farlex, Inc. | Feedback | For webmasters