Effects of Coupling Agents on the Oxidation and Darkening of Cellulosic Materials Used as Reinforcements for Thermoplastic Matrices in Composites.

J. MARTINEZ URREAGA [1][*]

M.C. MATIAS [1]

M.U. DE LA ORDEN [2]

M.A. LECHUGA MUNGUIA [3]

C. GONZALEZ SANCHEZ [4]

Oxidation and darkening occur during the processing of composites made from thermoplastic matrices and cellulosic reinforcements. We have studied the effects of several coupling agents on both the oxidation and darkening of cellulosic materials at temperatures close to those used in the processing of cellulose-reinforced thermoplastics. A maleated polypropylene wax (Epolene [E-43.sup.TM]) and two silanes (N-2-aminoethyl-3-aminopropyltrimethoxysilane and methyltrimethoxysilane) were used as coupling agents. Oxidation was measured by Diffuse Reflectance Infrared Fourier Transform (DRIFT) spectroscopy. Standard colorimetry was used to measure darkening. Coupling agent effects depend on the nature and extent of cellulose modification achieved by treatments and the nature of the coupling agent. Epolene wax E-43 produced scarce effects on both the oxidation and darkening of cellulosic materials at 200[degrees]C. Only for longer oxidation times was an increase in oxidation and darkening observed in E-43-treated sam ples. Silane coupling agents inhibited the formation of carbonyl and carboxyl groups for shorter oxidation times. The diaminosilane produced a stronger darkening, probably due to a chemical reaction that generated new chromophores containing C=N bonds.

INTRODUCTION

Lignocellulosic materials have some significant advantages as reinforcements in polymer composites, such as low density, biodegradability, low cost, availability of renewable natural resources, and good mechanical properties [1-16]. These advantages are especially interesting when low-cost, widely used thermoplastics (like polyethylene (PE) or polypropylene (PP)) are considered as matrices.

However, cellulose-thermoplastic composites present some disadvantages. One of them is the lack of compatibility between the polar, hydrophilic cellulose and the nonpolar, hydrophobic thermoplastic. Hence, several physical [9, 14, 17] and chemical [7, 9] methods involving the modification of cellulose and/or polymer surface have been tested in recent years in order to enhance fiber-matrix adhesion. Among the chemical methods, those involving the use of coupling agents have improved some mechanical properties of the composites.

Maleic anhydride (MA) and maleated polypropylenes (MAPP, graft copolymers of polypropylene and maleic anhydride) have been widely used as coupling agents. Both MA and MAPP may be covalently bonded to cellulose through an esterification process [7, 12, 18, 19]. Other chemicals, such as isocyanates [20-22] and silanes [2, 7, 22-27] of general formula [RSiX.sub.3] (where X is an easily hydrolyzable group and R is the organic group that provides the desired functionality) have also been proposed as coupling agents for these cellulose-reinforced thermoplastic composites.

Another disadvantage of cellulosic materials as reinforcements for thermoplastic matrices comes from their poor thermal stability [28, 29]. Cellulosic materials are in general easily oxidized at temperatures close to those used in the processing of thermoplastics. Severe degradation of cellulosic reinforcements may occur during the compounding of composite materials. Possible chemical changes in cellulosic material at these temperatures include dehydration, the formation of carbonyl and carboxyl groups, and a reduction of the degree of polymerization by bond scission, among others [29-31]. Such oxidation and degradation processes may have opposite effects on material properties and applications.

A moderate oxidation during the processing of cellulose-reinforced thermoplastics may have beneficial effects on the ultimate properties of the material, as it may improve fiber-matrix adhesion. In fact, several physical methods used for improving adhesion in these materials, such as flame or corona treatments, as well as some chemical methods, such as ozone treatment, include oxidation of the cellulose and/or the polymer surface. Such beneficial effects have been reported by some authors. Sapieha et al. [29] studied the relationship between strength properties and oxidation in composites made of untreated cellulose fibers and linear low density polyethylene. They reported that strength increased with processing time and temperature. This increase in strength was accompanied by the appearance of new infrared absorption bands at 1718 and 1735 [cm.sup.-1], due to oxidation.

On the other hand, several authors [15, 23, 27, 29] have reported a considerable darkening of cellulose-thermoplastic composites during compounding. A large amount of darkening, which may be due to a degradation process of the cellulosic reinforcement, constitutes a severe disadvantage for these materials, as it reduces their scope of application. For instance, darkening reduces the possibility of substituting these cheaper composite materials for existing polymers in applications that, for marketing purposes, require making goods in a wide variety of colors (e.g., white, red, or green). The intensity of darkening strongly depends on the processing time and temperature, as well as the type of fibers used [27, 29]. Sapieha et al. reported that a thermomechanical pulp produced composites with dark brownish color at 150[degrees]C, whereas composites prepared with alpha cellulose were only light brown when the processing temperature was 240[degrees]C [29]. The same authors also found that the degree of darkening was dependent on the water content in the cellulosic reinforcement: fibers with higher water content produced darker composites.

These reported results reveal that the darkening (and also oxidation) of cellulose-reinforced thermoplastics strongly depends on the chemical structure of the cellulosic reinforcement. Coupling agents used to improve the compatibility between cellulosic materials and thermoplastic matrices can modify the surface of cellulosic materials as well as the composite interfaces. Hence, coupling agents can alter significantly the darkening and oxidation characteristics of cellulosic materials and composites made from them. However, no published results about the influence of coupling agents on oxidation and darkening are known by the authors.

Our goals in this work were to characterize the oxidation and darkening of two cellulosic materials at 200[degrees]C, a temperature commonly encountered during the melt-blending and processing of cellulose-reinforced thermoplastic composites, as well as to measure the effects of several coupling agents on both the oxidation and darkening of cellulosic materials.

A cellulosic by product (from a Kraft pulp mill facility) and a cellulose filter paper were used as model cellulosic materials. The former is a typical representative of a cellulosic material used for making cellulose-reinforced composites, and the latter is a purified and bleached cellulose used as a control for the oxidation and darkening processes. A maleated polypropylene (Epolene [E-43.sup.TM]) and two silanes were used as coupling agents. N-2-aminoethyl-3-aminopropyltrimethoxysilane (AAPS) is a diamino silane primarily used in glass fiber-reinforced composites [32]. Methyltrimethoxysilane (MS) is primarily used to make inorganic surfaces hydrophobic. In a previous work [7] we studied the nature and extent of the reaction of the cellulosic byproduct with E-43 and the two silane coupling agents. Both E-43 and the silanes reacted with the cellulosic material, but the degree of modification was very different in each case. Only a limited esterification of the material surface was obtained when the maleated polypropylene was used. However, both silanes, when applied in the presence of organic peroxides as free radical sources blocked most of the cellulose reactive groups and were covalently bonded to the cellulosic material.

The experimental techniques used in this work were Diffuse Reflectance Fourier Infrared Transform (DRIFT) spectroscopy and standard colorimetry. DRIFT spectroscopy was used to measure the effects of the cellulose treatments with coupling agents on the cellulose oxidation at 200[degrees]C. Cellulose darkening under heating at 200[degrees]C and the effect of coupling agents on the darkening were measured as total color differences between heated and unheated samples.

MATERIALS AND METHODS

Cellulosic Materials

The cellulosic byproduct used in this work was obtained from CEASA. a Kraft pulp mill located in Navia, Asturias (Spain). It is a byproduct obtained in the manufacture of Kraft cellulose pulp, consisting of unbleached cellulose. Its cellulose content is higher than 95%, most of the rest being unreacted wood. Before treatment with coupling agents, large fragments of unreacted wood were eliminated. The material was washed with distilled water to ensure removal of water solubles from the surface and then dried in air at 65[degrees]C for 18 h to remove excess moisture. Dried powder was ground and sieved in order to isolate a fraction ranging from 0.25 to 0.5 mm that was used in all experiments. The other cellulosic material used in this work was Whatman No. 42 ashless filter paper. Its cellulose content was higher than 99% (ash yield 0.01%). This cellulose was also dried at 65[degrees]C before treatment with coupling agents.

After washing and drying, untreated materials were characterized by DRIFT spectroscopy, as previously described [7]. Both cellulosic materials present the characteristic infrared absorptions of cellulose. The main difference is a weak band at 1600 [cm.sup.-1] in the byproduct spectrum (see Figs. 1 and 2). This band may be assigned to C-C aromatic stretching vibrations, thus revealing the presence of a low amount of lignin in the byproduct. However, other compounds in the byproduct can contribute also to this absorption, such as some initially existing chromophores containing carbon-carbon double bonds conjugated to carbonyl groups. Infrared characterization also showed that drying in air at 65[degrees] did not cause any significant cellulose oxidation.

Coupling Agents

The maleated polypropylene used in this work was Epolene wax [E-43.sup.TM] (Eastman Chemical Products, Inc.). The Acid Number is 47. The [M.sub.n] and [M.sub.w] are 3900 and 9100, respectively. The E-43 was activated before use by heating at 160[degrees]C for 5 min. Dow Coming Iberica S.A. supplied N-2-aminoethyl-3-aminopropyltrimethoxysilane, AAPS ([([CH.sub.3]O).sub.3]Si-[CH.sub.2]-[CH.sub.2]-[CH.sub.2]-NH-[CH.sub. 2]-[CH.sub.2]-[NH.sub.2], Z-6020) and methyltrimethoxysilane, MS ([([CH.sub.3]O).sub.3][SiCH.sub.3], Z-6070). They were used as received.

Other Reagents and Solvents

Benzoyl peroxide (Merck) and acetone (Fluka) were used without further purification. Toluene (Fluka) was dried over [CaCl.sub.2].

Treatment With Coupling Agents

Treatments of cellulosic materials with coupling agents were carried out as previously described [7] for the study of cellulose modification with the same coupling agents. The use of the same reaction conditions allowed us to know the degree of cellulose modification achieved by these treatments.

For treatment with silanes, the cellulosic byproduct was reacted with each silane and benzoyl peroxide in acetone-water (95-5; v/v) mixtures, under nitrogen atmosphere at 60[degrees]C. Peroxide and silane concentrations were respectively 0.3 and 5% (based on cellulose weight). Cellulosic materials were treated with the maleated polypropylene using the method of Felix and Gatenholm [12]. The reaction was carried out in refluxing toluene for 5 min. The E-43 concentration was 5% (based on cellulose weight).

In all cases, treated samples were washed with the solvent and oven dried at 65[degrees]0 for 18 h before the oxidation and darkening were studied.

Oxidation and Darkening

Samples of the cellulosic byproduct (in powder form) were oven heated at 200[degrees]C. After the desired heating time, the oxidation and darkening were measured. The same procedure was used for the cellulose filter paper. In this case measurements were made on 2.5 cm by 4 cm sheets.

Diffuse Reflectance FTIR Spectroscopy

The DRIFT spectra of cellulosic materials before and after heating were obtained with a Mattson 3020 FTIR spectrometer, using a Specac diffuse reflectance accessory. Each spectrum was recorded at a resolution of 4 [cm.sup.-1], with a total of 90 scans. Background scans were obtained using high-purity KBr powder. Cellulosic samples were not diluted in KBr before collecting their spectra. All DRIFT spectra were plotted according to the Kubelka-Munk function [33,34].

Color Measurement

Color coordinates [L.sup.*], [a.sup.*], [b.sup.*], as defined by the Commission International d'Eclairage (CIE 1976 color space) were measured for heated and unheated samples using a Microcolor Dr. Lange colorimeter. A white standard was used to calibrate the colorimeter. Total color differences between heated and unheated samples, determined as

[delta][E.sup.*] = [[[([delta][L.sup.*]).sup.2] + [([delta][a.sup.*]).sup.2] [([delta][b.sup.*]).sup.2]].sup.1/2] (1)

were used to measure darkening of each sample. Each color difference reported in this paper is the average of five measurements.

RESULTS AND DISCUSSION

Oxidation of Cellulosic Reinforcements: Coupling Agent Effects

The oxidation of cellulosic materials during the processing of cellulose-reinforced thermoplastic composites is a relevant concern, as it is known that some oxidation at the composite interface results in improved mechanical properties of final composites [29]. Coupling agents, used to improve compatibility between some thermoplastic matrices and cellulosic reinforcements, may affect cellulose oxidation. In order to obtain information about these effects, we first characterized the oxidation of untreated cellulosic materials by DRIFT spectroscopy. Several samples of the cellulosic byproduct used in this work were subjected to heating in air at 200[degrees]C to study the oxidation of the material. Figure 1 shows the infrared spectra recorded after heating the byproduct for 0, 20, 50 and 90 min. Differences observed in these spectra reveal the main structural changes in the material.

The decrease in absorbance at 1640 [cm.sup.-1] when the heating time is increased may be related to a decrease in the water content, as the 1640 [cm.sup.-1] band has been assigned to adsorbed water in cellulosic materials [29, 35]. More interesting are the decrease in absorbance at 1600 [cm.sup.-1], revealing the consumption of initially existing unsaturated or aromatic compounds, and the appearance of new absorption bands in the 1700-1750 [cm.sup.-1] region, as a consequence of heating. For shorter heating times a broad absorption band centered at 1730 [cm.sup.-1] develops on heating. For longer times, a new overlapping band at 1710 [cm.sup.-1] becomes apparent. A similar result was reported by Sapieha et al [29]. They found two new absorption bands at 1735 and 1718 [cm.sup.-1] in infrared spectra (recorded in the transmission mode) of composites made of untreated cellulose fibers and linear low density polyethylene. An absorption band at 1735 [cm.sup.-1] appeared at low processing temperatures, whereas abs orption at 1718 [cm.sup.-1] appeared at higher temperatures.

Both 1730 and 1710 [cm.sup.-1] bands may be assigned to carbonyl stretching vibrations, thus revealing oxidation of the cellulosic byproduct. The 1730 [cm.sup.-1] absorption band appearing for shorter heating times may be assigned to carboxyl groups in oxidized celluloses on the basis of the work of Hurtubise and Krassig [36], and Forziati et al. [37]. The 1710 [cm.sup.-1] absorption band reveals the contribution to the global oxidation band of other carbonyl groups in oxidized celluloses, as for example, ketone or aldehyde groups.

Similar results were obtained when cellulose (What man paper) was subjected to heating at 200[degrees]C for 0, 20 and 90 min. The corresponding infrared spectra are shown in Fig. 2. A band at 1730 [cm.sup.-1], due to carboxyl groups, appears as the main result of oxidation for shorter heating times. In this case, the 1710 [cm.sup.-1] band is not so clearly noticeable, but the oxidation band maximum shifts from 1730 [cm.sup.-1] (for shorter heating times) to 1720 [cm.sup.-1] (for longer times), revealing the contribution of a new absorption at lower wave numbers. We may conclude that, for both cellulosic materials, the oxidation band initially centered at 1730 [cm.sup.-1] must be assigned to both carboxyl and other carbonyl groups existing in oxidized cellulose.

The effects of coupling agents on cellulose oxidation were studied by comparing the evolution of the normalized absorption at 1730 [cm.sup.-1] for treated and untreated materials. The cellulose infrared absorption band at 900 [cm.sup.-1], assigned to the [beta]-glycosidic linkages [35, 38], remained unchanged during the first stages of oxidation at 200[degrees]C and was used as an internal reference in order to normalize the results. The normalized absorption at 1730 [cm.sup.-1] was calculated as follows:

Absorption at 1730 [cm.sup.-1] = [([A.sub.1730]/[A.sub.900]).sub.t] - [([A.sub.1730]/[A.sub.900]).sub.t=0] (2)

where: [A.sub.1730] = Area under the 1730 [cm.sup.-1] band.

[A.sub.900] = Area under the 900 [cm.sup.-1] band.

[([A.sub.1730]/[A.sub.900]).sub.t=0] = Normalized absorption at time = 0 (before the heating process). This correction allows elimination of the contribution of other bands to the peak area at 1730 [cm.sup.-1].

Figures 3 and 4 illustrate plots of the normalized absorption at 1730 [cm.sup.-1] versus the oxidation time for untreated samples and samples treated with E-43 or silane coupling agents. The oxidation plots for untreated and E-43-treated paper samples are straight lines in the first stages of oxidation. This indicates zero-order kinetics. However, the oxidation plots for the untreated byproduct, as well as the silane-treated materials, do not correspond to zero-order kinetics. Figures 3 and 4 show that treatments with AAPS strongly modify the oxidation of the cellulosic byproduct, whereas treatments with E-43 only slightly affect the oxidation of cellulosic materials for shorter heating times. Only for longer heating times does a contribution of E-43 treatment to cellulose oxidation appear.

The main effect of cellulose treatment with AAPS is an inhibition of the oxidation process for shorter heating times (Fig. 4). This effect must be a consequence of cellulose modification achieved by the treatment. In a previous work we have shown that AAPS, when applied in the presence of organic peroxides as radical sources, is covalently bonded to cellulosic materials [7]. The inhibition effect of AAPS appears to confirm that this silane coupling agent, when applied in the presence of radical sources, effectively blocks most of the easiest oxidizable groups of cellulosic materials. The other silane used in this work, MS. also inhibites oxidation of the cellulosic byproduct for shorter heating times.

The different effect of treatments with E-43 on cellulose oxidation (Fig. 3) may be explained as a result of the different nature and extent of cellulose modification achieved with this treatment when compared with the modification obtained in cellulose treatments with silane coupling agents. We have shown that E-43, in refluxing toluene, reacts with some hydroxyl groups in the cellulosic byproduct, with the formation of ester bonds [7]. A similar result was obtained when the reaction of Whatman paper with E-43 in hot toluene was studied by DRIFT spectroscopy, in good agreement with results reported by Felix and Gatenholm [12]. The scant effect of E-43 on cellulose oxidation reveals that only a very reduced fraction of cellulose reactive groups are converted in the treatment with this maleated polypropylene wax, probably due to the steric hindrance of the relatively large polypropylene chains of E-43. Other authors have reported also that only a limited esterification of cellulosic materials may be achieved with maleated polypropylenes [12, 18, 39].

Another relevant question in the oxidation of treated cellulosic reinforcements is the possible transformation of the organic functionalities introduced by treatments with coupling agents. Again, the observed behavior was clearly dependent on the nature of the coupling agent. Infrared spectra of E-43-treated cellulosic materials subjected to oxidation were very similar to those corresponding to oxidized untreated samples. No specific oxidation (or transformation) of the maleated polypropylene units was detected under heating at 200[degrees]C. The same behavior was observed for MS when the infrared spectra of oxidized samples of an MS-treated cellulosic byproduct were analyzed.

However, [NH.sub.2] groups of AAPS were easily degraded under the same reaction conditions. Figure 5 shows infrared spectra of the AAPS-treated cellulosic byproduct before and after heating in air at 200[degrees]C for 60 min. Spectrum 5a, corresponding to the unheated material, shows the characteristic absorption bands of AAPS attached to the cellulosic material [7] at 1550 and 1600 [cm.sup.-1] (amine deformation band). Both characteristic absorptions disappear almost totally after heating the sample for 60 min. (spectrum 5b). The difference spectrum (5b-5a, curve 5c) shows a new band at 1655 [cm.sup.-1] besides the negative bands at 1550 and 1600 [cm.sup.-1] and the 1725 [cm.sup.-1] band assigned to the carbonyl and carboxyl groups produced in the oxidation. The new band at 1655 [cm.sup.-1] in the infrared spectrum of oxidized samples may be assigned to C=N bonds (stretching vibrations) [40]. These results reveal that the diaminosilane attached to cellulosic materials suffers severe degradation when heated at temperatures close to those used in the melt-blending and processing of cellulose reinforced thermoplastics. This degradation should be taken into account while analyzing the effects of this coupling agent on the properties of the thermoplastic-matrix composite material.

Darkening of Cellulosic Reinforcements: Coupling Agent Effects

Composites made from some thermoplastic matrices and lignocellulosic reinforcements are often strongly colored [15, 23, 27, 29]. Some factors are known to account for the darkening, such as the processing time and temperature, or cellulose water content [29]. Darkening may be related a priori to some degradation of the cellulosic reinforcement during the composite processing. We have studied the effects of treatments with coupling agents on the darkening of some cellulosic materials.

First we measured the effect of treatments on the color of cellulosic materials before heating. Table 1 shows the color coordinates [L.sup.*] [a.sup.*] [b.sup.*] (CIE 1976) of treated and untreated samples. Our cellulosic byproduct (unbleached) was initially more colored than the Whatman paper, revealing the presence of chromophores in the starting byproduct. Some of these initially existing chromophores would be partially responsible for the absorption band at 1600 [cm.sup.-1] (assigned to unsaturated compounds) observed in the infrared spectrum of the by-product before heating. Table I also shows that the treatments used in this work do not produce any significant darkening of the cellulosic materials.

In order to obtain information about the effect of coupling agents on the darkening after heating cellulosic materials, we measured the color development (as color differences, [delta][E.sup.*]) for treated and untreated samples subjected to oxidation in air at 200 [degrees]C. The main results obtained are shown in Figs. 6 and 7. Both cellulosic materials suffer a substantial darkening under such reaction conditions. This darkening on heating can be attributed to the presence of chromophores formed from the oxidative degradation of one or more components of cellulosic materials. As shown above in the infrared study of oxidation, during oxidation of both cellulosic materials, carbonyl and carboxyl groups are formed from hydroxyl groups. The formation of certain carbonyl groups in the anhydroglucose units has been shown to account for the yellowing of celluloses [28, 41].

Figure 6 shows that Whatman paper suffers a faster darkening than our byproduct at 200 [degrees]C. This appears to be an amazing result, as Figs. 3 and 4 show that the byproduct is oxidized faster than the Whatman paper under the same conditions. However, it should be considered that Figs. 3 and 4 show the evolution of the absorption at 1730 [cm.sup.-1], and both carbonyl and carboxyl groups contribute to this absorption.

Moreover, it should be considered that the cellulosic byproduct was initially more colored than the Whatman paper. As mentioned above, some initial chromophores could be compounds with carbonyl groups conjugated to carbon-carbon double bonds. In the oxidation of the byproduct in air, some initially existing chromophores could be degraded, including the oxidation of colored carbonylic compounds to colorless carboxylic compounds. This oxidation of initial chromophores would account for the decreasing in absorption at 1600 [cm.sup.-1] observed when the byproduct was subjected to oxidation (see Fig. 1). The carboxyl groups formed in such oxidation would also contribute to the absorption at 1730 [cm.sup.-1] observed in the infrared spectrum of the oxidized byproduct.

On the other hand, our byproduct appears to contain some lignin. It is known that pulps that contain large amounts of lignin tend to yellow faster [41]. However, the infrared study has shown that the lignin content of our cellulosic byproduct is minimal. A negligible effect of this lignin on material oxidation and darkening should then be expected.

Figure 6 also shows the effect of E-43 treatments on cellulose discoloration rate under heating in air. For shorter heating times modification with E-43 has a scant effect on the darkening of both cellulosic materials. A negative effect may be observed for longer heating times. These results, which are similar to those observed in the infrared analysis of E-43 effects on oxidation, appear to confirm that only a limited chemical modification of the cellulosic materials is achieved in treatments with maleated polypropylenes. Moreover, these results reveal that the degradation of maleated polypropylene units deposited onto the cellulosic materials does not contribute significantly to the darkening of treated materials, at least for shorter heating times at moderate temperatures.

In contrast, treatment with AAPS causes a pronounced effect on darkening of the cellulosic byproduct. Figure 7 shows that darkening is faster for AAPS-treated samples than for untreated samples. This negative effect should be assigned to the organofunctional group of AAPS (-[([CH.sub.2]).sub.3]-NH-[([CH.sub.2]).sub.2]-[NH.sub.2]) and not to the organic peroxides used as a radical source in treatment, as treatment with MS in the same reaction conditions does not cause such a pronounced effect on the darkening rate (see also Fig. 7).

This effect of AAPS treatment on darkening should be analyzed in conjunction with observed effects on oxidation (see Fig. 4). As AAPS treatment produces an inhibition effect on the appearance of the absorption band at 1730 [cm.sup.-1] (due to carbonyl and carboxyl groups), we can conclude that the chromophores responsible for the darkening of AAPS-treated samples are different from those responsible for the darkening of untreated or E-43-treated samples. Infrared analysis also shows that amine groups of AAPS disappear in the first stages of oxidation.

Darkening produced by AAPS treatment should be assigned to new chromophores formed from a chemical reaction (taking place during oxidation at 200[degrees]C) between amine groups of AAPS and one or more components of the cellulosic material (or compounds formed from the oxidative degradation of such components). Several chemical reactions between amines and carbohydrates, involving the formation of colored compounds, may occur at 200[degrees]C. Amines react with carbonyl compounds to form imines. A well-known similar reaction takes place between some diamines, such as phenylhydrazine, and compounds such as glucose or anhydroglucose. Osazones produced in this reaction are yellow compounds and contain C=N bonds [42, 43]. The formation of similar colored compounds in the degradation of AAPS-treated cellulosic materials would explain the observation of a new band at 1655 [cm.sup.-1] (assigned to C=N bonds), as well as the disappearance of amine bands in the infrared spectrum of the AAPS-treated byproduct after pa rtial oxidation (see Fig. 5).

This evidence of formation of new chromophores during oxidation of some treated samples indicates that the coupling agent must be properly selected in order to avoid undesirable darkening processes during the melt-blending and processing of composites made from cellulosic reinforcements, coupling agents, and thermoplastic matrices.

CONCLUSIONS

We have studied the effects of several coupling agents on the oxidation and darkening of two different cellulosic materials subjected to heating in air at 200[degrees]C. The effects on oxidation are relevant, as cellulose oxidation plays a significant role in compatibilizing the cellulosic reinforcements with the thermoplastic matrices. The effects on darkening are also of importance, because a large darkening of cellulose-reinforced thermoplastics during compounding reduces their usefulness in making goods with a wide variety of colors, a severe limitation in industry. The major conclusions of this study are:

* Two silane coupling agents (N-2-aminoethyl-3-aminopropyltrimethoxysilane (AAPS) and methyltrimethoxysilane (AAPS) have been shown to produce significant effects on the oxidation and darkening of cellulosic materials subjected to heating at temperatures close to those used during the melt-blending and processing of cellulose-reinforced thermoplastic composites. The effects of each coupling agent have been related to the nature and extent of the chemical modification of the cellulose induced by the coupling agent.

* A maleated polypropylene wax (Epolene E-43) produces scarce effects on both the oxidation and darkening of cellulosic materials at 200[degrees]C. Only for longer heating times was an increase in darkening and oxidation observed in E-43-treated samples.

* For shorter heating times, silane coupling agents inhibit the formation of carbonyl and carboxyl groups during the oxidation of cellulosic materials. This result may be explained by considering that silane coupling agents convert most of the cellulose reactive groups.

* The effect of silane coupling agents on cellulose darkening strongly depends on the nature of the silane organofunctional group. A diaminosilane (AAPS) produces a stronger discoloration of treated samples, probably due to a chemical reaction that generates new chromophores containing C=N bonds.

(1.) Departamento de Ingenieria Quimica Industrial E. T. S. I. Industriales Universidad Politecnica de Madrid Jose Gutierrez Abascal 2, 28006-Madrid, Spain

(2.) Departamento de Quimica Organica I, E. U. Optica Universidad Complutense de Madrid Arcos de Jalon, S/N, 28037-Madrid, Spain

(3.) Instituto de Ciencias Exactas Universidad Autonoma del Estado de Hidalgo. Cd. Universitaria Carr. Pachuca-Tulancingo, Km 4,5, 42074-Pachuca, Hidalgo, Mexico

(4.) Departamento de Ingenieria Quimica y Tecnologia del Media Ambiente Universidad de Oviedo Julian Claveria, S/N, 33071-Oviedo, Spain

(*.) To whom correspondence should be addressed. Fax: [phone]: E-mail: [e-mail]

REFERENCES

(1.) R. T. Woodhams, G. Thomas, and D. K. Rodgers, Polym. Eng. Sci., 24, 1166 (1984).

(2.) L. M. Matuana, J. J. Balatinecz, and C. B. Park, Polym. Eng. Sci., 38, 765 (1998).

(3.) L. M. Matuana, C. B. Park, and J. J. Balatinecz, Polym. Eng. Sci., 37, 1137 (1997).

(4.) B. D. Park and J. J. Balatinecz, Polym. Compos., 18, 79 (1997).

(5.) A. K. Rana, A. Mandal, B. C. Mitra, R. Jacobson, R. Rowell, and A. N. Banerjee, J. Appl. Polym. Set, 69. 329 (1998).

(6.) C. Clemons, A. J. Giacomin, and J. A. Koutsky, Polym. Eng. Sci., 37, 1012 (1997).

(7.) M. C. Matias, M. U. de la Orden, C. Gonzalez Sanchez, and J. Martinez Urreaga, submitted.

(8.) K. Oksman and C. Clemons, J. Appl. Polym. Sci., 67, 1503 (1998).

(9.) D. Maldas and B. V. Kokta, Composite Interfaces, 1, 87 (1993).

(10.) P. Zadorecki and A. J. Michell, Polym. Compos., 10, 69 (1989).

(11.) J. M. Felix, P. Gatenholm, and H. P. Schreiber, Polym. Compos., 14, 449 (1993).

(12.) J. M. Felix and P. Gatenholm, J. Appl. Polym. Sci., 42, 609 (1991).

(13.) S. Takase and N. Shiraishi, J. Appl. Polym. Sci, 37, 645 (1989).

(14.) A. K. Bledzki, S. Reihname, and J. Gassan, J. Appl. Polym. Sci., 59, 1329 (1996).

(15.) G. E. Myers, I. S. Chahyadi, C. Gonzalez, C. A. Coberly, and D. S. Ermer, Intern. J. Polymeric Mater., 15, 171 (1991).

(16.) C. Gonzalez, C. M. Clemons, G. E. Myers, and T. M. Harten, Mat. Res. Soc. Symp. Proc., 266, 127(1992).

(17.) S. Doug, S. Sapieha, and H. P. Schreiber, Polym. Eng. Sci., 33, 343 (1993).

(18.) M. Kazayawoko, J. J. Balatinecz, and R T. Woodhams, J. Appl. Sci., 66, 1163 (1997).

(19.) N. E. Marcovich, M. M. Reboredo, and M. I. Aranguren, J. Appl. Polym Sci., 61, 119(1996).

(20.) D. Maldas and B. V. Kokta, Polym. Eng. Sci., 31. 1351 (1991).

(21.) R. G. Raj, B. V. Kokta, D. Maldas, and C. Daneault, J. Appl. Polym. Sci., 37, 1089 (1989).

(22.) D. Maldas, B. V. Kokta, and C. Daneault, J. Appl. Polym. Sci., 37, 751 (1989).

(23.) M. Xanthos. Plast. Rubber Process. Appl., 3, 223 (1983).

(24.) A. Beshay and S. V. Hoa, Sci. Eng. Compos. Mater., 2, 85 (1992).

(25.) M. H. Schneider and K. I. Brebner, Wood Sci. Technol., 19, 67 (1985).

(26.) P. Bataille, L. Ricard, and S. Sapieha, Polym. Compos., 10, 103 (1989).

(27.) F. M. B. Coutinho, T. H. S. Costa, and D. L. Carvalho, J. Appl. Polym. Sci., 65, 1227 (1997).

(28.) M. Lewin, Macromol, Symp., 118, 715 (1997).

(29.) S. Sapieha, J. F. Pupa, and H. P. Schreiber, J. Appl. Polym. Sci, 37, 233 (1989).

(30.) C. F. Cullis, D. Goring, and M. M. Hirschler, in Cellulose and Its Derivatives. Chemistry, Biochemistry and Applications, p. 401, J. F. Kennedy, G. O. Phillips. D. J. Wedlock, and P. A. Williams, eds., Ellis Horwood. Chichester, United Kingdom (1985).

(31.) A. Blazej and M. Kosik, in Cellulose and Its Derivatives. Chemistry. Biochemistry and Applications, p. 97, J. F. Kennedy, G. O. Phillips, D. J. Wedlock, and P. A. Williams, eds., Ellis Horwood. Chichester, United Kingdom (1985).

(32.) A. T. Dibenedetto and P. J. Lex, Polym. Eng. Sci., 29, 543 (1989).

(33.) K. C. Cole, D. Noel and J. J. Hechler, Polym. Compos., 9, 395 (1988).

(34.) J. A. Schmidt and C. Heitner, Tappi J., 76, 117 (1993).

(35.) R. K. Samal and M. C. Ray, J. Appl. Polym. Sci., 64, 2119 (1997).

(36.) F. G. Hurtubise and H. Krassig, Anal. Chem., 32, 177 (1951).

(37.) F. H. Forziati, J. W. Rowen, and E. K. Plyler, J. Res. Natl. Bur. Stan., 46, 288 (1951).

(38.) T. Kondo and C. Sawatari, Polymer, 37, 393 (1996).

(39.) C. Joly, R. Gauthier, and M. Escoubes, J. Appl. Polym. Sci, 61, 57 (1996).

(40.) R. T. Conley. Espectroscopia Infrarroja, Alhambra, Madrid (1979).

(41.) H. A. Carter, J. Chem. Educ., 73, 1068 (1996).

(42.) N. L. Allinger, M. P. Cava, D. C. De Jongh, C. R. Johnson, N. A. Lebel, and C. L. Stevens, Organic Chemistry, p. 705, Worth Publishers, New York (1971).

(43.) A. Streitwieser and C. H. Heathcock, Quimica Organica, p. 988, Interamericana, Mexico (1986).

                Initial Color Coordinates (CIE LAB 1976) of
                Treated and Untreated Cellulosic Materials.
                 [L.sup.*] [a.sup.*] [b.sup.*]
Paper              96.5       0.1       1.4
Paper + E-43       95.8       1.0      -0.6
Byproduct          60.3       2.6       8.6
Byproduct + E-43   60.8       3.4       8.2
Byproduct + AAPS   59.8       3.3      10.5
Byproduct + MS     62.8       3.7      10.3