Surface characteristics of untreated and modified hemp fibers.

INTRODUCTION

Natural cellulosic fibers are extensively used as reinforcing material in polymer substrates due to their high specific properties, low cost, unlimited availability, and renewable nature. Composites manufactured using them find applications in diverse fields like auto components, building materials, and furniture [1, 2]. However, poor interfacial adhesion between fibers and synthetic resins leads to a decline in mechanical properties of the composites [3-4].

Role of Acid-Base Interactions in Adhesion

The surface energy of a material can be described by the sum of dispersive component and specific interactions. Dispersive component refers to London dispersion forces, and specific interactions refer to the polar, ionic, electrical, metallic, and acid-base interactions. Fowkes and Mostafa proposed that dispersion forces and acid-base interactions are the primary forces operating across the interface [5]. Thus, work of adhesion can be written as

[W.sub.a] = [W.sub.a.sup.d] + [W.sub.a.sup.AB]

where [W.sub.a], [W.sub.a.sup.d], and [W.sub.a.sup.AB] are the total work of adhesion, work of adhesion due to dispersion forces and work of adhesion because of acid-base interactions, respectively. Donnet and coworkers also showed that acid-base interactions are strongly correlated to interfacial shear strength carried out on single fibers and interlaminar shear strength of the composites [6]. Moreover, acid-base interactions are also the useful chemical interactions available for modification [7]. Hence, to design new modification methods for improving fiber-matrix adhesion and meaningful interpretation of the existing ones, quantitative determination of surface acid-base characteristics of natural fibers is important.

The objective of this study is to determine dispersive component of surface energy and acid-base characteristics of hemp fibers, using inverse gas chromatography (IGC), Modification methods like alkalization, acetylation, coupling agents, and corona treatment have been reported to improve the interfacial adhesion [3, 8, 9]. But according to authors' knowledge, changes in dispersion component of surface energy and acid-base character of various natural fibers because of these surface treatments are not fully explored. Therefore, an attempt has been made to determine them. Alkalization and acetylation were chosen because of their wide use in the field of cellulose chemistry [10]. Acid-base interactions between fibers and unsaturated polyester matrix were also determined using specific interaction parameter as defined by Schultz et al. [11]. Finally, changes in specific interaction parameter were compared with the improvement in the flexural properties of resin transfer molded (RTM) composites.

Inverse Gas Chromatography

IGC is a useful technique to measure surface characteristics of different types of materials. Good reviews are available in the literature [12-13]. In IGC at infinite dilution, solid material is used as the stationary phase. A small amount of volatile probes is injected into the column, and their interaction with the stationary phase is studied using retention volume ([V.sub.n]), which is defined as the amount of carrier gas required to elute a volatile probe from the column [14]. The relationship used is

[V.sub.n] = F([T.sub.r] - [T.sub.o])

where, F is the corrected flow rate: [T.sub.r], the retention time of a particular probe; and [T.sub.o], the retention time of the noninteracting probes (generally methane or air).

Molar-free energy of adsorption ([DELTA]G) is related to net retention volume by the following relation.

[DELTA]G = RTln([V.sub.n]) + C (1)

where, R is the gas constant; T, the temperature; and the value of C depends on the reference state.

According to Schultz et al. free energy of adsorption is related to work of adhesion by following relation [11].

[DELTA]G = Na[W.sub.a] (2)

where, N is the Avogadro's number; a, the surface area of a single probe; and [W.sub.a], the work of adhesion.

Dispersive Interactions

According to Fowkes, work of adhesion between two materials interacting only due to dispersion forces is given by [15]

[W.sub.a.sup.d] = 2[square root of [[gamma].sub.l.sup.d][[gamma].sub.s.sup.d]] (3)

where [[gamma].sub.l.sup.d] and [[gamma].sub.s.sup.d] are the dispersive components of surface free energy of the interacting solid and liquid, respectively. Combining Eqs. (1)-(3), we get

RTln([V.sub.n]) = 2Na [square root of [[gamma].sub.l.sup.d][[gamma].sub.s.sup.d]] + C.

A plot of RTln([V.sub.n]) versus a [square root of [[gamma].sub.l.sup.d]] should yield a straight line, in case of probes interacting only due to dispersion component of surface energy. From the slope of the straight line [[gamma].sub.s.sup.d] can be calculated.

Acid-base Interactions

In the case of polar probes, both dispersion and acid-base interactions occur between probes and adsorbent. Therefore, the net retention volume measured using IGC is due to both of these interactions. However, Fowkes showed that both these interactions act independent of each other [15], i.e., net retention volume for polar probes is the sum of contribution due to each of the two components. To determine the contribution of acid-base interactions to the total, the contribution of dispersion interactions has to be separated. To achieve this, it is assumed that the dispersion component of polar probes is equal to that of alkanes of comparable size. Therefore, for polar probes, contribution of the acid-base interactions can be determined by subtracting the free energy of adsorption as estimated from alkane line from the global free energy of adsorption.

The free energy of adsorption ([DELTA][G.sup.AB]) is related to enthalpy of adsorption ([DELTA][H.sup.AB]) by

[DELTA][G.sup.AB] = [DELTA][H.sup.AB] - T[DELTA][S.sup.AB]

where, [DELTA][S.sup.AB] is the entropy of adsorption due to acid-base interactions.

A plot of [DELTA][G.sup.AB] versus T (temperature) should result in a straight line with intercept equal to [DELTA][H.sup.AB]. According to Flour and Papirer [16]

[DELTA][H.sup.AB] = [K.sub.A]DN + [K.sub.B]AN

where DN and AN are donor number and acceptor number as defined by Guttmann [17], [K.sub.A] and [K.sub.B] describe the acidic and basic character of the fibers [11].

Therefore, a plot of [DELTA][H.sup.AB]/AN versus DN/AN is expected to be linear with the slope and the intercept equal to [K.sub.A] and [K.sub.B] respectively.

According to Schultz et al. specific interaction parameter [11], I, for acid-base interactions can be defined as

I = [K.sub.A.sup.f][K.sub.B.sup.m] + [K.sub.A.sup.m][K.sub.B.sup.f]

[FIGURE 1 OMITTED]

where, superscript f and m refer to fiber and matrix, respectively. This parameter can be used to determine the acid-base interactions occurring between fibers and the resin.

EXPERIMENTAL

Materials

Hemp fibers used in this experiment were supplied by Hempline, Canada. The polymer used was unsaturated polyester resin, supplied by Progress Plastics (Stypol 040-8086). Methyl ethyl ketone peroxide was procured from Sigma Aldrich. The analytical reagents used were n-hexane, n-heptane, n-octane, n-nonane, chloroform, ethyl acetate, ethyl ether, tetrahydrofuran, acetone, sodium hydroxide, glacial acetic acid, and acetic anhydride. Characteristics of different probes used for IGC analysis are shown in Table 1.

Alkali Treatment

Hemp fibers were soaked in 6% w/v sodium hydroxide solution at room temperature for 48 h. Fibers are removed, thoroughly washed with distilled water, and dried at room temperature.

Acetylation

Hemp fibers were treated with glacial acetic acid at room temperature for 1 h. Fibers were removed and further treated with acetic anhydride containing few drops of concentrated sulfuric acid (catalyst) for 5 min. Treated fibers were then washed thoroughly with distilled water and dried at room temperature.

Column Preparation and IGC Procedure

IGC measurements were done with Perkin-Elmer 8500 gas chromatograph fitted with flame ionization detector. To ensure flash vaporization, the injection port was kept at least 50 K above the boiling point of the probes. A column of length 30 cm and internal diameter of 4 mm was filled with 4 g of hemp fibers. Helium was used as the carrier gas. Corrected flow rate of helium varied from 30 to 55 ml/min. Small quantities of probes were injected into the column using Hamilton syringes. Peaks were found to be symmetrical and independent of the amount injected. Average of five measurements was taken to calculate retention volume, with methane as the marker.

Fiber Mat and Composite Manufacturing

Randomly oriented hemp fibers were carefully spread on a woven mat of glass fibers. Then it is covered with another sheet of woven glass fiber, and pressed in a hot press at a temperature of 313 K for 30 min. Hybrid composites with glass fiber mat at the top and the bottom, and hemp fibers in the middle are manufactured with untreated, acetylated, and alkalized hemp fibers separately using RTM. Unsaturated polyester is used as the matrix material: 25 wt% of hemp fibers and 11 wt% of glass fibers are maintained in all the composites. The RTM procedure is described in one of the earlier article from our group [18].

Mechanical Testing

The flexural strength and flexural modulus are measured using ASTM D790 standards.

RESULTS AND DISCUSSION

Dispersive Component

[FIGURE 2 OMITTED]

The dispersive component of untreated, alkalized, and acetylated hemp fibers was calculated from plots of RTln([V.sub.n]) versus a [square root of [[gamma].sub.l.sup.d]]. Dispersive component of resin is also calculated similarly. Figure 1 shows plot for untreated fibers at 313 K. The linear relationship in n-alkanes illustrates that this technique works well in case of natural fibers. The values of dispersive component of surface energy ([[gamma].sub.s.sup.d]) at different temperatures are summarized in Table 2. Alkalization and acetylation increased the dispersive component of hemp fiber. This is possibly due to dissolution of low-energy surface impurities and surface exposure of relatively higher energy cellulose.

Acid-Base Interactions

Free energy of adsorption ([DELTA][G.sup.AB]) was plotted against temperature (T) for all the probes in all the four cases, i.e., untreated, alkalized, acetylated hemp fibers, and resin. Enthalpy of adsorption ([DELTA][H.sup.AB]) for each probe in all four cases was calculated from these plots. One of such plot is shown in Figure 2. Finally, [DELTA][H.sup.AB]/AN was plotted against DN/AN for each case as shown in Figure 3, and values of [K.sub.A] and [K.sub.B] were determined from it. These values are were summarized in Table 2. Hemp fibers are found to be basic, which is probably due to the presence of extractives like triglycerides, which exhibit basic behavior. Alkalization cleans the fiber surface by dissolving extractives and hemicellulose. This is expected to reduce the basic character of the surface and increase acidic character due to exposure of cellulose, which is predominantly acidic [20]. Acetylation was found to slightly decrease the basic character of hemp fibers. This is because dissolution of extractives is expected to decrease basicity whereas acetylation leads to an increase in basicity due to esterification of the accessible hydroxyl groups in the cell wall [21]. Magnitude of basicity imparted by acetylation depends upon the extent of the reaction. The unsaturated polyester resin is also found to be basic.

[FIGURE 3 OMITTED]

Values of specific interaction parameter, as defined by Schultz et al. [11], were calculated for each type of fiber and resin combination. From Table 2 it is clear that acid-base interactions with unsaturated polyester increased in alkalized and acetylated fibers.

Correlation with Mechanical Properties of the Composites

Composites manufactured with alkalized and acetylated fibers showed improved flexural properties. Values of flexural strength and flexural modulus are shown in Table 3. Highest improvement in acid-base interactions between fiber and matrix is observed in case of alkali-treated fibers and the same is true for the mechanical properties of the composites. This shows the importance of acid-base interactions in describing the interfacial properties of natural fiber reinforced composites.

CONCLUSIONS

* IGC at infinite dilution has proven to be a convenient tool for measurement of surface energy and acid-base characteristics of natural fibers. Changes in final properties of the composites due to various modification methods can also be explained using this technique.

* Alkalization and acetylation make the hemp fiber amphoteric but basicity predominates in case of acetylated fibers due to esterification of hydroxyl groups.

REFERENCES

1. T.G. Schuh, Renewable Materials for Automotive Applications, Daimler-Chrysler AG, Stuttgart (1999).

2. L.T. Drzal, A.K. Mohanty, and M. Mishra, National Science Foundation for Advancing Technologies in Housing (NSF-PATH) (2001).

3. L.Y. Mwaikambo and M.P. Ansell, Die Angew. Makromol. Chem., 272, 108 (1999).

4. J.M. Flex and P. Gatenholm, J. Appl. Polym. Sci., 42, 609 (1991).

5. F.M. Fowkes and M.A. Mostafa, Ind. Eng. Chem. Prod. Res. Dev., 17, 3 (1978).

6. S.J. Park and J.B. Donnet, J. Colloid Interface Sci., 206, 29 (1998).

7. D.W. Dwight, F.M. Fowkes, D.A. Cole, M.J. Kulp, P.J. Sabat, L. Salvati, and T.C. Huang, J. Adhes. Sci. Technol., 4, 619 (1990).

8. J. Gassan and A.K. Bledzki, Prog. Polym. Sci., 24, 221 (1999).

9. M.N. Belgacem, P. Bataille, and S. Sapieha, J. Appl. Polym. Sci., 53, 379 (1994).

10. D. Klemm, B. Philipp, T. Heinze, U. Heinze, and W. Wagenknecht, Comprehensive Cellulose Chemistry, Vols. 1 and 2, Wiley-VCH, Weinheim (1998).

11. J. Schultz, L. Lavielle, and C. Martin, J. Adhes., 23, 45 (1987).

12. P. Mukhopadyay and H.P. Schreiber, Colloid Surfaces A: Physicochem. Eng. Aspects, 100, 47 (1995).

13. D.R. Lloyd, T.C. Ward, and H.P. Schreiber, ACS Symposium Series 391, American Chemical Society, Washington, DC (1989).

14. J.R. Conder and C.L. Young, Physiochemical Measurement by Gas Chromatography, Wiley, New York, 25 (1979).

15. F.M. Fowkes, Ind. Eng: Chem., 56, 40 (1964).

16. C.S. Flour and E. Papirer, Ind. Eng. Chem. Prod. Res. Dev., 21, 337 (1982).

17. V. Guttmann, The Donor Acceptor Approach to Molecular Interactions, Plenum Press, New York (1983).

18. D. Rouison, M. Sain, and M. Couturier, Comp. Sci. Technol., 64, 629 (2004).

19. G.M. Dorris and D.G. Gray, J. Colloid Interface Sci., 71. 93 (1979).

20. J.C. Berg and P.N. Jacob, Langmuir, 10, 3086 (1994).

21. R.M. Rowell, Mol. Cryst. Liq. Cryst., 418, 153 (2004).

Deepaksh Gulati

Department of Chemical Engineering and Applied Chemistry, University of Toronto, 200 College St., Toronto, Ontario M5S 3E5, Canada

M. Sain

Faculty of Forestry, University of Toronto, 33 Willcocks St., Toronto, Ontario M5S 3B3, Canada

Correspondence to: M. Sain; e-mail: [e-mail]

Contract grant sponsor: AUTO21 Network of Centres of Excellence.

TABLE 1. Characteristics of the different probes used in the analysis
[11, 17].

                 Area                  [[gamma].sub.l.sup.d]
Probe            ([[Angstrom].sup.2])  (mJ/[m.sup.2])         DN

Hexane           51.5                  18.4                   --
Heptane          57                    20.3                   --
Octane           62.8                  21.3                   --
Nonane           68.9                  22.7                   --
Chloroform       44                    25.9                    0
Ethyl Acetate    48                    19.6                   17.1
Ethyl Ether      47                    15                     19.2
Tetrahydrofuran  45                    22.5                   20
Acetone          42.5                  16.5                   17

Probe            AN    Character

Hexane           --    --
Heptane          --    --
Octane           --    --
Nonane           --    --
Chloroform       23.1  Acidic
Ethyl Acetate     9.3  Amphoteric
Ethyl Ether       3.9  Basic
Tetrahydrofuran   8    Basic
Acetone          12.5  Amphoteric

TABLE 2. Values of dispersion component of surface free energy at
different temperatures, acid-base characteristics, and specific
interaction parameter.

                       [[gamma].sub.s.sup.d] (mJ/[m.sup.2])
                       313  333  353  373
Type                   K    K    K    K           [K.sub.A] (au)

Untreated fibers       38   35   29   25          0.05
Alkali treated fibers  43   39   33   30          0.19
Acetylated fibers      41   36   30   28          0.12
Polyester resin        40   32   22   15          0.18
Cellulose              48 [+ or -] 3 at 293 K [19]

                                       I =
                                       [K.sub.A.sup.f][K.sub.B.sup.m]+
Type                   [K.sub.B] (au)  [K.sub.A.sup.m][K.sub.B.sup.f]

Untreated fibers       0.24            0.070
Alkali treated fibers  0.16            0.129
Acetylated fibers      0.20            0.100
Polyester resin        0.53            --
Cellulose

TABLE 3. Effect of treatments on flexural strength and flexural modulus
of the composites.

Type of composite      Flexural strength   Flexural modulus

Untreated fibers       6.66 [+ or -] 0.22  106.08 [+ or -] 3.85
Alkali treated fibers  7.75 [+ or -] 0.14  124.14 [+ or -] 7.42
Acetylated fibers      6.91 [+ or -] 0.30  117.96 [+ or -] 3.90