Clays for polymeric nanocomposites.
As the pace of nanocomposites research and application has intensified, it became apparent that the conventional approach to understand and optimize the polymeric nanocomposites (PNC) performance is unsatisfactory. The misconception is that, it is simply a case of scale and that technology applicable to discontinuous reinforced composites is transferable directly to PNC. However, at the nanoscale, the conventional approach used to determine structure-property relationships is unsuitable when the length-scale of the reinforcement and that of the onset of nonbulk (localized) properties coincide.
The nanosized particles are characterized by large surface-to-volume ratio, for example, the specific surface area of layered montmorillonite (MMT) is [A.sub.sp] = 750-800 [m.sup.2]/g, thus about half of atoms are on the surface. These two quantities are smaller for spherical particles, for example, a 10 nm sphere with similar density has [A.sub.sp] [approximately equal to] 230 [m.sup.2]/g and 15% of surface atoms. In consequence, the layered clay platelets are capable of immobilizing large quantity of polymeric segments, thus manifold increasing effective volume fraction of the reinforcing solids. Thus, in comparison to classical polymer matrix composites (PMCs), the filler content needed for enhancing material performance of PNCs is significantly reduced to, typically, 1-5 vol%. The addition of nanoparticlcs has a beneficial effect on a wide suite of mechanical and physical properties (e.g., stiffness, strength, thermal stability, fire retardancy, and fracture toughness), improving functionality to levels not achievable using larger scale fillers in the same quantities.
PNC performance and the functionality increases of the matrix polymer by the addition of nanoparticles depend on a number of factors including shape, size, size distribution, chemical composition, and purity of the nano-filler. These aspects are of prime interest to the Versailles Project on Advanced Materials and Standards (VAMAS), Technical Work Area on Polymer nanocompo-sites (TWA-33). The relevance of the work cannot he understated particularly as the potential global market for PNCs is expected to exceed [pounds sterling]lbn/annum by 2015. and it hinges on the availability of accurate and reliable techniques for measuring the above properties. Material property enhancement and their modeling depend on the availability of improved measurement methods and reliable information of the PNC individual components.
As this article summarizes work performed within the TWA-33. a short historical note is appropriate. In 1982. leaders of the G7 countries created the VAMAS for supporting trade through the international collaborative projects aimed at providing the technical basis for harmonized measurements, testing, specifications, and standards for advanced materials. VAMAS carries out its work within topical groups known as the Technical Working Areas (TWA's) [see www.vamas.org]. The TWA-33 is relatively recent, viz., www.vamas.org/twa33/documents/2009_vamas_twa33_general.pdf. Its aim is the selection of specific test methods for PNCs, generation of round robin test results, and transmitting the technical data to the international standardization organization (ISO) for establishing new standards. The aim of Project #1 is the characterization of nanoparticles, whereas that of Projects #2 and #3 is development of test methods for the electrical, dielectric, and mechanical properties of PNCs.
This article summarizes the early results of Project #1 generated by the members listed in Table 1. The authors welcome the reader's comments and encourage joining the organization by contacting any member listed in that Table.
TABLE 1. VAMAS TW-33 Project #1 participants. Country Name Organization Brazil C. A. Achete, Institute Nacional de B. Archanjo, Metrologia, E. Gravina, Normalizacao e A. Kuznetsov Qualidade industrial--INMETRO Brazil L. Hecker de Carvalho Universidade Federal de Campina Grande Canada F. Perrin-Sarazin, Nation Research Council Canada, IMI. L. A. Utracki Boucherville, QC Italy G. Camino Politecnico di Torino. Sede di Alessandria Japan M. Okamoto Toyota Technological Institute, Nagoya Mexico N. Gonzalez-Rojano Centra Nacional de Metrologia, Queretaro, J. L. Cabrera Qt UK W. Broughton National Physical Laboratory, Teddinglon South Africa L. Adlem Nation Metrology Institute, NMISA Country Name Address TWA-33 Brazil C. A. Achete, firstname.lastname@example.org; Member B. Archanjo, email@example.com: Member E. Gravina, firstname.lastname@example.org; Member A. Kuznetsov email@example.com; Member Brazil L. Hecker de firstname.lastname@example.org Member Carvalho Canada F. Perrin-Sarazin, Florence.Perrin@cnrc-nrc.gc.ca; Member L. A. Utracki email@example.com Chair Italy G. Camino firstname.lastname@example.org Co-chair Japan M. Okamoto email@example.com Current Project Leader Mexico N. firstname.lastname@example.org Member Gonzalez-Rojano J. L. Cabrera Member UK W. Broughton email@example.com Member South L. Adlem firstname.lastname@example.org Member Africa
The study used sodium salts of commercial clays: (1) natural MMT from Southern Clay Products, Cloisite[R]-[Na.sup.+] (C-[Na.sup.+]); (2) a semisynthetic fluoro-hectorite, Somasif ME-100 (ME-100) from CBC (Japan); and (3) synthetic fluoro-tetrasilicic mica from Topy Industries (Topy-[Na.sup.+]).
The C-[Na.sup.+] is a Wyoming clay that contains 4-9 wt% moisture and 7 wt% loss on ignition (LOI) (1). It is an off-white powder with the specific and bulk density [rho] = 2.86 and 0.34 g/mL, respectively. In the "as supplied" powder, the aggregated particle diameter ranges from about 2 to 13 [micro]m. The X-ray diffraction (XRD) interlayer spacing is [d.sub.001] = 1.17 nm and the unit cell composition is [[Al.sub.3.34] [Mg.sub.0.66] [Na.sub.0.66]]([Si.sub.8][O.sub.20])[(OH).sub.4]. The C-[Na.sup.+] platelets thickness is, [t.sub.z] = 0.96 nm. the cation exchange capacity is CEC = 0.92 meq/g, and the average nominal aspect ratio is p [equivalent to] (diameter/thickness) [approximately equal to] 280.
Tateyama et al. patent (2)describes the synthesis of ME-100, which involves the following steps: mixing the powders of purified, natural talc with that of sodium fiuoro silicate, [Na.sub.2][SiF.sub.6], and lithium fluorite, LiF, then heating it in an electric furnace for several hours at 850-900[degrees]C. The reaction results in high aspect ratio lamellar phyllosilicate, with a structure similar to hectorite. The unit cell composition is [(NaF).sub.[2.2]][([MgF.sub.2]).sub.[0.1]][(MgO).sub.[5.4]][([SiO.sub.2]).sub.8]. The particle size of the agglomerated bright white powder ranges from about 5 to 7 [micro]m, its specific surface area is 9 [m.sup.2] /g, the nominal aspect ratio is p [less then or equal to] 6000, density [rho] = 2.6. CEC = 1.2 meq/ g, and the interlayer spacing is [d.sub.001] = 0.95 nm (3), (4)
During the preliminary tests, synthetic clay was also examined. Since the company does not provide sodium salt, a small quantity of Topy-[Na.sup.+] was prepared at NRCC/IMI (5). Topy products are high purity materials manufactured from salts and oxides at the temperature above 1500[degrees]C followed by controlled crystallization. The expandable fluoro-mica is ion-exchanged with ammonium salts. Typieally. its unit cell composition is [NaMg.sub.[2.5]]-[Si.sub.4][O.sub.10][F.sub.2] The analysis gives Na = 4-9, Li < 0.5, MgO = 21-29, [SiO.sub.2] = 55-65, and F = 6-15 wt%. The aspect ratio of Topy-[Na.sup.+] is p [less then or equal to] 5000, CEC = 0.80 meq/g, and [d.sub.001] = 1.23 nm (4).
The clays have been characterized for:
1. Shape, size, and size distribution of platelets:
2. Chemical composition;
Shape, Size, and Size Distribution of Clay Platelets (6). The objective of this task is determination of the shape of individual clay platelets by measuring their orthogonal size, viz. platelet length (the longest dimension = L), platelet width (perpendicular to length = W) and the platelet thickness (orthogonal to the clay platelet surface = [t.sub.z]), followed by calculation of their averages and size distributions. Evidently, success of the process critically depends on the extent of clay dispersion, that is, on exfoliation. Accordingly, suspension of 0.002 g/L of Na-clay in deminc-ralized water is prepared by preswelling the clay, and then dispersing the platelets in ultrasonic bath at 40 kHz and T = 60-80[degrees]C. A droplet of the resulting suspension is deposited on a polycarbonate (PC) membrane filter (SPI-Pore[TM], pore diameter = 220 nm). After drying, the membrane with clay platelets is observed in scanning (SEM), transmission (TEM). or atomic force (AFM) microscope.
For reliable average measures of the platelets (L, W, and thickness, [t.sub.z]), over 200 particles (from 30-40 micrographs) need to be analyzed. When using SEM with the field-emission gun (FEG-SEM) a drop of exfoliated clay suspension was deposited on the PC-filter, for good contrast metalized with Pt and/or Au. and observed under low voltage (e.g., 1 kV). Next, the platelet contours were manually traced and scanned for the image analysis using commercial software (Visilog, SigmaScan Pro. Image-Pro[R]-Plus, etc.).
When using AFM, a drop of exfoliated clay suspension was deposited on a hot, freshly cleaved mica flake, and scanned using a Si cantilever tip at a speed of 2-29 mm/s in a 300 kHz tapping mode. The size and size distribution was computed using the instrument software.
Chemical Composition of Clay Platelets The energy-dispersive X-ray spectroscopy (SEM-EDX) provides information on the chemical composition of a specimen. The high-energy primary electrons in SEM eject electrons from the specimen's atoms. Replacements of the inner shell electrons by those from the outer ones engender X-rays with the energies characteristic of the elements. The signal originates from the few micrometers thick surface layer. As clay may have locally diverse composition, each particle should be sampled from [greater then or equal to]5 locations, collecting [greater then or equal to]30 scans for statistical analysis. The instrument must be calibrated using known specimen atomic composition at the test voltage.
Few particles of clay powder on metallic support are covered with thin layer of colloidal graphite. After drying. the specimen is placed in SEM for observation of its shape and possible presence of impurities. Next, the ele- mental analysis is carried out at accelerating voltage of 15-35 kV and about X40k magnification. The specimen spectrum is corrected for the method artifacts and carbon presence (left after graphitization) and then the statistics of chemical composition is computed.
Clay Impurities. Commercial clays contain diverse and variable impurities. For example, the recent publication reports that commercial bentonite contains 63 wt% of sodium (MMT-Na ) with the reminder consisting of contaminants, such as quartz, kaolinite, carbonates, etc. (7), (8). As the natural mineral composition varies not only with the geographical location, but also with strata, it is important that for reproducible performance of PNC the properties of ingredients are constant. There are three types of impurities: organic (e.g., humic substances, HS), nonexpandable clays (e.g., amorphous clays, vermiculite. kaolin) and diverse particulate minerals (e.g., silica, feldspar, gypsum, orthoclase, apatite, halite, calcite, dolomite, quartz, biotite. muscovite. chlorite. hematite) (8).
For the analysis of contaminants, 0.5 g clay was dispersed in 50 mL demineralized water (9), (10). The suspension was centrifuged. sedimenting particulate minerals, aggregates and non-expandable clays. After drying at 60[degrees]C, the sediment was analyzed by XRD. The supernatant suspension was filtered through a cellulose filter producing well aligned in z-axis, several micrometer thick-layered clay films. Dry film was subjected to four XRD tests at ambient temperature: (1) directly, (2) after impregnation with ethylene glycol (overnight at 60[degrees]C), (3) after heating to 400[degrees]C, and after heating to 550[degrees]C.
Size, Size Distribution and Shape of the Clay Platelets
SEM Results. As an example, Fig. 1 shows three SEM micrographs at the same magnification and manually traced contours of clay platelets. The top two images display an exfoliated platelet and a short stack of ME-100 reported by Carvalho et al. (11), whereas in the bottom one there are 16 distinguishable platelets of C-[Na.sup.+] described by Perrin-Sarazin and Sepehr (6). The platelets' contours (in Fig. ID) were used for the statistical analysis by Image-Pro[R] -Plus. Such micrograph image analysis (MIA) yields the number and weight averages of L, W (i.e., [L.sub.n], [L.sub.w] [W.sub.n], [W.sub.w]) and their ratios [[(L/W).sub.n], [(L/W).sub.w]] listed in Table 2.
[FIGURE 1 OMITTED]
TABLE 2. Statistical analysis of three clays . Length L (nm) Width, W (nm) Clay Counted [L.sub.n] [L.sub.w] [W.sub.n] [W.sub.w] platelets C-[Na.sup.+] 234 290 350 183 219 ME-100 304 872 1097 572 743 Topy-[Na.sup.+] 447 1204 1704 761 1186 Error - Ratios Clay [(L/W).sub.n] [(L/W).sub.w] C-[Na.sup.+] 1.58 1.60 ME-100 1.52 1.48 Topy-[Na.sup.+] 1.58 1.44 Error [+ or -] 0.2 [+ or -] 0.2
Figure 2 presents images of the three sodium clays: C-[Na.sup.+] ME-100. and Topy-[Na.sup.+]. Since the micrograph pore diameter is the same (220 nm) evidently, the platelet size dramatically increases from C-[Na.sup.+] to Topy-[Na.sup.+] (6).
[FIGURE 2 OMITTED]
Table 3 summarizes the C-[Na.sup.+] and ME-100 clay platelets dimensions determined by the National Physical Laboratory (NPL) (12). Figure 3 shows these data plotted in probability coordinates. Apparently, while C-[Na.sup.+] has smaller clay platelets and Gaussian (normal) size distribution, ME-100 has larger platelets and more complex distribution, with higher than normal population of smallest and largest plates. However, as shown in Fig. 3B, the ratio of L/W for both clays is similar, with the mean value of [(L/W).sub.av] = 1.4 [+ or -] 0.4..
[FIGURE 3 OMITTED]
TABLE 3. Statistics of 204 clay platelets dimensions of C-[Na.sup.+] and ME-100 . Clay Parameter Mean Std Minimum Median Maximum Dev C-[Na.sup.+] Length (nm) 543 144 182 534 1129 Width (nm) 377 100 147 377 891 Perimeter (nm) 1786 472 569 1795 4190 Area (X [10.sup.3] 148 73 27 142 734 [mm.sup.2]) Effective D (nm) 211 50 93 213 483 Length/Width 1.48 0.38 1.00 1.42 4.14 ME-100 Length (nm) 811 244 338 783 1557 Width (nm) 589 198 209 547 1340 Perimeter (nm) 2735 813 994 2607 5640 Area (X [10.sup.3] 340 88 55 299 1240 [mm.sup.2]) Effective D (nm) 317 88 132 309 628 Length/Width 1.44 0.44 1.00 1.29 4.22
AFM Results. Several Project members used AFM for determining the shape and size of clay platelets. As for the SEM measurements, here also full exfoliation is essential, thus the preparation of specimens is similar, viz. clay concentration of 0.002 g/L in demineralized water prepared by ultrasonication at about 40 kHz and T= 60-80[degrees]C; a drop of the suspension should be deposited on hot, freshly cleaved mica. Figure 4 displays few examples (6), (10). The number of micrographs taken for C-[Na.sup.+] and ME-100 clays was 160 and 92, respectively (10). Due to the high clay concentration, the automatic image analysis was unable to distinguish individual platelets. However, manual examination of the automatic scans yielded correct orthogonal dimensions, for example, for C-[Na.sup.+] and ME-100 the clay platelet thickness (see Fig. 4), [t.sub.z] = 1.2 and 1.0 [+ or -] 0.2 nm, length: L = 222 and 640 nm, respectively. It is noteworthy that some micrographs (e.g., see Fig. 5) evidence the presence of contaminating particles, clearly visible in the C-[Na.sup.+]
[FIGURE 4 OMITTED]
[FIGURE 5 OMITTED]
Other Methods. Some participants also evaluated the automatic particle size analyzers, including highly advanced ones, with rather disappointing result. For example, the Horiba Partica is the second-generation Low Angle Light Scattering (LALS) instrument. It uses the Mie Scattering for measuring the particle size within the range of 0.01-3000 [micro]m, with stability and accuracy guaranteed to 0.6%, precision to within 0.1%. The sample-to-sample measurement time is 60 s. The system has a centrifugal pump and a 130 W in-line ultrasonic probe for dispersion that "allows the complete sample dispersion and analysis sequence to be handled without the need for external sample preparation" (13).
For testing clays, first a 5 wt% clay suspension was ball-milled then kept at 85[degrees]C for 12 h. For comparison, low-concentration clay suspensions (0.5 and 0.005 wt%) were also prepared by 1 h ultrasonication and then kept for 12 h at 85[degrees]C. The results for C-[Na.sup.+] and ME-100 are shown in Figs. 6A and B, respectively-for comparison the platelet size distribution determined by SEM with MIA is also displayed (14). In the case of C-[Na.sup.+], depending on concentration, two or three peaks are detectable, with the first ones located not far from that of MIA, viz. [L.sub.peak] = 209, 261, and 350 nm for MIA, 0.05 and 5 wt% LALS. The other peaks with [L.sub.peak] = 1.9 to 20.1 [micro]m reflect the presence of aggregates - their size increases with clay content.
[FIGURE 6 OMITTED]
Figure 6B shows that for ME-100, there is a significant difference in the locations of MIA and LALS peaks. The peak maximum of the former is at 0.60 [micro]m, whereas the first peaks from LALS are located at [L.sub.peak] = 7.3, 7.3, and 1 1.2 [micro]m for 0.005, 0.05, and 5 wt% clay, respectively. Interestingly, the data indicate that even at the highest clay content of 5 wt%, the ball milling eliminated the largest ME-100 aggregates, evident in the ultrasonicatcd samples at peak positions [L.sub.peak] = 945 and 712 [micro]n, for 0.005 and 0.05 wt% clay, respectively. It is noteworthy that the nominal maximum size of ME-100 platelets is about 6 [micro]n (3). Similar results were obtained by Gonzalcz-Rojano and coworkers (Table 4) using the dynamic light scattering (DLS) or photon correlation spectroscopy (PCS) (10).
TABLE 4. Results of the PCS clay size determination . Clay Cone. Av. Std (nm) PI (-) Sid (-) 2-nd peak (g/L) Dia. ([micro]m) (nm) C-[Na.sup.+] 0.002 270 30 0.305 0.05 5 C-[Na.sup.+] 0.20 338 55 0.276 0.08 5 ME-100 0.002 469 66 0.504 0.08 5 ME-100 0.20 474 82 0.347 0.15 5 Cone. = concentration of clay in suspension; Av. Dia = average diameter of clay particles; Std = standard deviation; PI = dimensionless polydispersity index - a measure of nanoparticles size distribution; 2-nd peak = size of aggregates at the peak value.
The lack of agreement between AFM and DLS may be related to the flow orientation of the large ME-100 clay platelets. Zanetti-Ramos el al. measured D of spherical polyurethane particles. AFM indicated presence of two generations of particles with D = 127 [+ or -] 26 and 218 [+ or -] 36 nm, whereas DLS showed single generation with [D.sub.av] = 262 nm (15).
Chemical Composition of Clays
The chemical composition of C-[Na.sup.+] and ME-100 was determined following the earlier described procedure. For example, the X-ray spectra were obtained using Oxford Instruments Inca Energy EDX system with a Si(Li) detector. Table 5 summarizes the results from four laboratories, with statistics of 30 scans at NPL (16) and Politecnico di Torino (17), The composition was adjusted by eliminating data for carbon and normalizing the content to 100 wt%.
TABLE 5. Chemical composition of C-[Na.sup.+] and ME-100 (6), (10), (15), (17). All result in weight percent. Clay Parameter O Na Mg C-[Na.sup.+] Mean (6) 52.40 2.51 2.65 C-[Na.sup.+] Mean (10) 50.16 3.60 1.34 C-[Na.sup.+] Mean (17) 64 [+ or -] 4 3 [+ or -] 0 - C-[Na.sup.+] Mean (16) 48.21 2.38 1.21 Std. deviation 7.38 0.31 0.26 Max. 63.57 3.01 1.76 Min. 31.20 1.79 0.82 Clay Parameter Al Si Fe Total C-[Na.sup.+] Mean (6) 11.78 30.66 - 100 C-[Na.sup.+] Mean (10) 11.78 29.41 3.71 100 C-[Na.sup.+] Mean (17) 11 [+ or -] 4 22 [+ or -] 3 - 100 C-[Na.sup.+] Mean (16) 11.05 32.10 5.04 100 Std. 1.43 4.30 2.63 deviation Max. 13.70 40.14 16.63 Min. 8.69 21.92 2.47 Clay Parameter O F Na ME-100 Mean (6) 27.04 28.90 3.88 ME-100 Mean (10) 37.37 19.92 5.40 ME-100 Mean (17) 51 [+ or -] 3 7 [+ or -] 2 3 [+ or -] 0 ME-100 Mean (15) 43.59 12.48 2.59 Std. deviation 3.77 3.92 0.49 Max. 49.60 18.79 3.73 Min. 34.97 4.12 1.34 Clay Parameter Mg Si Al Total ME-100 Mean (6) 16.44 23.74 - 100 ME-100 Mean (10) 14.32 22.44 0.56 100 ME-100 Mean (17) 15 [+ or -] 3 22 [+ or -] 2 - 100 ME-100 Mean (15) 14.50 26.84 - 100 Std. deviation 1.47 6.00 - Max. 17.16 42.59 - Min. 11.22 17.89 -
In Table 6, the converted data from Table 5 (with results reported by Sepehr et al. (18)) express the molar composition of crystalline cells. The literature formula (4), (8) and those based on results from the four participating laboratories are shown. As evident from the statistics, in spite of the large number of spectra the statistical standard error of measurements is significant, for C-[Na.sup.+] and ME-100 ranging from [+ or -] 13 to 53 and from [+ or -] 9 to 31 in wt%, respectively. It is noteworthy that results from other laboratories not always fall within this margin of error.
TABLE 6. C-[Na.sup.+] and ME-100 molar composition from EDS (8), (10), (16), (18). All results in weight percent. Source C-[Na.sup.+] ME-100 Literature [[Al.sub.3.34] [(NaF).sub.2.2] (8) [Mg.sub.0.66] [([MGf.sub.2]).sub.0.1] [Na.sub.0.66]] [(MgO).sub.5.4] [([SiO.sub.2]).sub.8] ([Si.sub.8] [O.sub.20]) [(OH).sub.4] NRC (18) [[Al.sub.3.2] [(NaF).sub.1.6] [Mg.sub.0.8] [([MgF.sub.2]).sub.6.4] [Na.sub.0.8]] [(MgO).sub.0][([SiO.sub.2]).sub.8] ([Si.sub.8] [O.sub.20]) [(OH).sub.4] CENAM (10) [[Al.sub.3.3] [(NaF).sub.2.3] [Fe.sub.0.5] [([MgF.sub.2]).sub.4.1] [Mg.sub.0.4] [(MgO).sub.1.8] [Na.sub.1.2]] [([Fe.sub.2][O.sub.3]).sub.0.1] ([Si.sub.8] [([SiO.sub.2]).sub.8] [O.sub.20]) [(OH).sub.4] NPL (16) [[Al.sub.2.9] [(NaF).sub.0.94] [Fe.sub.0.6] [([MgF.sub.2]).sub.2.3] [Mg.sub.0.35] [(MgO).sub.2.7] [([SiO.sub.2]).sub.8] [Na.sub.0.72]] ([Si.sub.8] [O.sub.20]) [(OH).sub.4] 0 62.61 [+ or -] 9.5 53.98 [+ or -] 4.7 ([+ or -] 9% ([+ or -] 15% error) error) Na 2.14 [+ or -] 0.27 2.22 [+ or -] 0.43 ([+ or -] 19% ([+ or -] 13% error) error) Mg 1.04 [+ or -] 0.21 1 1.84 [+ or -] 1.18 ([+ or -] 10% ([+ or -] 21% error) error) Al 8.61 [+ or -] 1.19 - ([+ or -] 13% error) Si 23.74 [+ or -] 3.26 18.94 [+ or -] 4.3 ([+ or -] 22% ([+ or -] 13% error) error) Fe 1.87 [+ or -] 0.98 - ([+ or -] 52% error) F - 13.02 [+ or -] 4.02 ([+ or -] 31% error)
In addition, for identification of impurities present in C-[Na.sup.+] and ME-100 clays qualitative elemental analysis using SEM-EDX and X-ray fluorescence (XRF) was conducted. These data will be presented in the next part (10).
As displayed in Fig. 5, AFM is capable detecting the presence of particulate contaminants (10). However, for the systematic impurity identification a more elaborate procedure (described in Test Procedures-Clay Impurities) is required. In principle, centrifugation of clay suspension separates the components into two fractions, the heavier sediments and the supernatant liquid suspension. The XRD identifies the components of solid powders. Figures 7-10 present the key results.
Figures 7 and 8 compare the XRD diffractograms of "as received" clays (C-[Na.sup.+] and ME-100, respectively) with their sediments spectra enriched with impurities. In Figs. 7a and b, the basal interlayer spacing of MMT is evident at [d.sub.001] = 1.18 nm. However, while the diffractogram Fig. 7a mainly shows the pattern of C-[Na.sup.+] that in 7b is significantly more crowded. After expanding its intensity scale the diffractogram (not shown) displays diffraction peaks corresponding to MMT (now [d.sub.001] = 1.35) as well as those of contaminants: quartz (dominant peak at d = 0.33 nm with 0.42, 0.23, and 0.18 nm secondary ones), cristobalite (d = 0.40 nm), anorthite (d = 0.62 nm), and analcime (d = 0.56 nm). Spacings of all these agree with XRD database lists. The absence of vermiculite peak at d = 0.154 nm is noteworthy.
[FIGURE 7 OMITTED]
[FIGURE 8 OMITTED]
Figure 8 provides the corresponding information to that in Fig. 7, but now for ME-100. As evident from the diffractogram 8a, ME-100 is nearly contaminants-free, with a basal plane spacing of fluorohectorite at [d.sub.001] = 1.24 nm. The XRD of powder, recovered from the clay sediment (8b), mainly repeats the pattern of the clay with traces of such contaminants as reclorite, [NaAl.sub.4][(SiAl).sub.8][O.sub.20][(OH).sub.4].[2H.sub.2]O with d = 2.40 nm. quartz (d= 0.33 nm), talc with d = 0.96 nm, and anthophyllite [[Mg.sub.7][Si.sub.8][O.sub.22][(OH).sub.2]]. While the rectorite presence indicates local overheating during ME-100 synthesis, the other minerals probably entered the synthesis with talc (10).
The supernatant liquid after centrifugation of C-[Na.sup.+]or ME-100 has formed aligned, thin film on the filter. Figures 9 and 10 display the XRD diffractograms of C-[Na.sup.+] and ME-100, respectively. The four diffractograms in each Figure represent, respectively:
[FIGURE 9 OMITTED]
[FIGURE 10 OMITTED]
1. Dry film sample
2. Clay (as in A) but preswollen in ethylene glycol
3. Clay heated to 400 [degrees]C
4. Clay heated to 550[degrees]C
It is noteworthy that spectra 9A and 10A indicate presence of a single dominant mineral-in the former Figure the peak at [d.sub.001] = 1.25 nm has a shoulder at [d.sub.001] =1.48 nm, whereas in the latter a single peak at [d.sub.001] = 1-24 nm is present. Auxiliary tests identified the presence of two smectites in C-[Na.sup.+] powder (9A): MMT and beidellite, and single clay in ME-100 powder (10A), the fluorohectorite. Thus, [d.sub.001] of C-[Na.sup.+] film expanded after treatment with ethylene glycol from the original value of 1.25 and 1.48 to 1.70 nm (9B), while that of ME-100 expanded from [d.sub.001] = 1.24 nm to 1.70 nm (10B). Heating these films at 400[degrees]C caused a collapse of the interlayer spacing to the theoretical value of [d.sub.001]-- 0.96 [+ or -] 0.01 nm (the smectite diffractograms Figs. 9C and 9D show a weak peak at 0.34 nm). Heating at 550[degrees]C produced no additional changes in XRD spectra. Table 7 provides a list of 14 minerals identified by the XRD method.
To confirm the identity of contaminants in C-[Na.sub.+] and ME-100, the centrifuged sediments was subjected to qualitative identification of chemical composition using SEM-EDX and XRF. The results may be summarized as follows:
1. In C-[Na.sub.+] suspension SEM-EDX detected Na, Mg, Al, Si, O, K, Fe as major and Ca. Ti. S. as minor components.
2. In C-[Na.sup.+] suspension XRF detected Na, Al, Si, Fe, as more intensive and Ti, K, S, Mg, O as less intensive peaks.
3. In ME-100 suspension SEM-EDX detected F, Mg, Si, O, as major and Na, Al as minor components.
4. In ME-100 suspension. XRF detected Si, Mg, as more intensive and Fe, Na, Al, O as less intensive peaks.
The elements detected in both analyses are in part from the sedimented impurities identified in the XRD measurements.
Natural clays originate from the hydrothermal alteration of alkaline volcanic ashes and rocks of the Cretaceous period (85-125 million years ago), deposited by winds in seas and lakes (19), (20). The plastics industry is interested in crystalline, swellable clays (e.g., hydrous silicates of Al and Mg, or phyllosilicates) with the specific surface area of about 750[m.sup.2]/g, which makes them highly physico- and chemico-sorptive. As a consequence, clay deposits not only contain admixed impurities (e.g., quartz, sand. silt, feldspars, mica, chlorite, opal, volcanic dust, fossil fragments, heavy minerals, sulfates, sulfides, carbonates, zeolites, and amorphous contaminants), but also 3-5 wt% of adsorbed organic and inorganic compounds.
Bentonite if the raw clay, usually containing 60-80 wt% of MMT. After purification and conversion to sodium-salt, the MMT-[Na.sup.+] crystalline unit cell is (8):
Layer sandwich of two Si-O [[[Al.sub.3.33] tetrahedron sheets and a central [Mg.sub.0.67]].sup.+0.67] AIMg-O octahedral sheet with 0.67 ([Si.sub.8] [O.sub.20]) negative charge per unit cell [(OH).sub.4] Aqueous interlamellar layer ([nH.sub.2]O) [Na.sup.+0.67] containing 0.67 [Na.sup.+] cations per unit cell
Purification of natural clays is a labor-intensive, complex process involving nearly 300 steps. There is extensive literature on the topic (8), (21-26). In an early report on purification of Wyoming clays Earley et al. outlined laboratory purification and fractionation by progressive sedimentation and centrifugation that eventually lead to a suspension of exfoliated individual platelets, still containing amorphous and crystalline impurities (27). Roberson et al. (28) also used sequential centrifugation method to separate individual MMT platelets from microaggregates constituting about 80 wt% of clay. Interestingly, the authors suggest that the aggregates are "due to the interlocking of flakes in microaggregates during crystal growth, which prevents their complete separation in dilute suspension." The tabulated length and width of the Wyoming clay platelets was 300-350 and 200-150 nm, respectively. The clay was brownish, containing 5-7 wt% [Fe.sub.2][O.sub.3] (28).
Because of the variability of natural clay composition and inevitable presence of contaminants, there is a growing tendency for its replacement by synthetic or semisynthetic clays, which contain no organic and much less inorganic impurities (4). The major obstacle for this replacement is the limited number of manufacturers and their low production capacity.
This article considers three aspects of clays deemed important for the production of reliable and reproducible PNC: (1) size and shape of clay platelets, (2) their chemical composition, and (3) presence of impurities. As evident from the data presented above, there is significant diversity of results obtained for the tested clays.
Shape and Size
Starting with the first aspect, the clay platelet shape and size was examined using three methods:
1. SEM followed by contour tracing of well-defined single platelets and then statistical evaluation of their polydispersity.
2. AFM followed by an automatic or manual image analysis.
3. Analyses of flowing suspensions.
Of the three methods, SEM and AFM followed by manual identification and analysis of platelet size offer reliable approach. In the case of AFM with automatic image analysis, the computed thickness ranged from I to 35 nm, indicating uncritical acceptance of any particle size and shape, viz. exfoliated platelets and large stacks. Similarly, as shown in Fig. 6, LALS does not distinguish exfoliated from nonexfoliated particles. Two obvious problems of the applied procedures are the inadequate exfoliation and irreproducible sampling that neglects the effects of the time and sedimentation.
The shape of clay platelets is irregular, but the length-to-width ratio was found to be nearly the same for all clays so far tested [(L/W).sub.av] = 1.4 [+ or -] 0.4. While the thickness of layered silicates is about [t.sub.z] [approximately equal to] 1 nm, the average inscribed (or effective) diameter, [D.sub.in], and the aspect ratio: p = [D.sub.in]/[t.sub.z], range from about 25 to 6000 (18). The natural clay, exemplified in this study by C-[Na.sup.+], has mid-size platelets, [D.sub.in] [approximately equal to] 211 nm (Table 3), and it follows the normal probability curve (in Fig. 3 the correlation coefficient r = 0.99).
Swellable 2:1 phyllosilicates have ionically imbalanced tetrahedral (T) or octahedral (O) layers that create a need for the formation of an interlayer space, which houses the balancing ions and moisture. The hectorite crystalline structure differs from that of talc by having some divalent [Mg.sub.2+] atoms in O-layer replaced by monovalent ones, [Li.sup.+] or [Na.sup.+] (4). Accordingly, the production of the semisynthetic ME-100 amounts to a partial replacement of Mg by Li or Na, accompanied by a partial substitution of -OH groups by -F (2). The process requires heating a mixture of talc with, for example, LiF and [Na.sub.2][SiF.sub.6]. For the control of clay crystalline structure, the furnace temperature is critical, namely, the fiuoro-hectorite produced at 700-750[degrees]C has too small interlayer spacing, [d.sub.001] = 0.91 nm, while that produced at 780-900[degrees]C the desired spacing: [d.sub.001] = 1.61 nm (2). However, at T > 950[degrees]C instead of layered hectorite a needle-like richterite, Na[NaCa][[(Mg,[Fe.sup.2+]).sub.5]([Si.sub.8][O.sub.22])[(OH,F).sub.2], forms (29). The natural talc contains such impurities as, for example, [Fe.sub.2][O.sub.3], [Al.sub.2][O sub.3], CaO, [AsS.sub.2], [NiS.sub.2], (30), thus its rigorous purification is essential for high quality of the semisynthetic ME-100.
Determination of the ME-100 platelet dimensions is more difficult than that of C-[Na.sup.+] as this clay does not exfoliate easily. This is because of factors like: (i) large clay platelets D [less than or equal to] 6 [micro]m, thus high probability of inter-plate crystalline welding (e.g., by insufficient substitution of [Mg.sub.2+] by a monovalent ion); (ii) large platelet size result in tendency for aggregation and sedimentation of its suspensions: (iii) partial replacement of -OH groups by -F reduces the hydrophilic clay character. It is noteworthy that the rate of molecular diffusion (which controls exfoliation) into stacks of circular discs with diameter [D.sub.in] decreases with [D.sub in.sup.2] (8).
The measured platelet length, L < 1.6 [micro] (Table 3, Fig. 3a), is significantly smaller than that quoted by the manufacturer, L [less than or equal to] 6 [micro]m. Tables 2 and 3 show similar values of the mean length, namely, [L.sub.n] = 872 and 811 [+ or -] 244 nm, respectively. This may suggest that the largest platelets either sedimented to the bottom of the test tube before the clay suspension was sampled or that they were fragmented during ultrasonication. Contrary to the nominal dimensions, the largest platelets were found to be those of Topy-[Na.sup.+].
Achieving full exfoliation of Na-clays poses a problem, smaller for C-[Na.sub.+] and Topy-[Na.sub.+] more serious for ME-100. In the patent literature, the first step of clay intercalation is its dispersion at concentration w [less than or equal to] 2 wt%. Thus, Na-clay is added to demineralized water at temperature of 50-80[degrees]C with vigorous stirring, ultrasonication or ball milling for at least 4 h. In industry, clay is considered exfoliated when at least 80% platelets are fully and randomly dispersed. The dispersed system undergoes centrifugation to remove contaminants and aggregates and then is treated with intercalating onium salts (8), (31), ( 32). It is noteworthy that as Fig. 6 shows. ultrasonication is less effective in eliminating aggregates than ball milling. However, there are reports that mechanical grinding changes the clay morphology toward amorphization (33). By contrast, as L. Perez-Maqueda et al. reported, ultrasonication of mica (20 kHz, 0.75 kW, t = 10-100 h) resulted in reduction of particle size by a factor of 10, while preserving the crystalline structure and yielding aggregate-free platelets with relatively narrow size distribution (34). It may be that the standard laboratory ultrasonic bath (42 kHz, 70 W output) that is used for the clay dispersion is not suitable for peeling clay platelets from the stack. Recent report of MMT sonication at 35 and 70 W for up to 60 min showed a progressive shift of the LALS bimodal signals toward a decade smaller diameters. For example, starting with [D.sub.0] [greater than or equal to] 200 nm platelets their diameter after sonication for 1 h was [D.sub.60] [greater than or equal to] 30 nm, with the lower peak position at D [approximately equal to] 70 nm, what may indicate attrition of exfoliated platelets (35). Cavitation during ultrasonication locally produces high temperatures, for example, 5075 [+ or -] 156 K (36).
The second aspect of clay suspension is its stability. Ideally, one would wish that only contaminants and aggregates formed by faulty crystallization would sediment, leaving stable suspension of exfoliated clay platelets for the analysis. The influence of acid-base balance and electrolyte content (pH and pK) on the stability of clay suspensions has been frequently discussed, both based on mathematical model or experimental data (21), (37), (38). The suspension stability depends on the repulsive interaction between clay double layers, controlled by the clay composition (Si-OH and Al-OH sites on the platelet faces and edges) pH and pK that may lead to edge-to-face, face-to-face or edge-to-edge association. The experiments indicated that for Wyoming clay the most stable system is at pH = 8.5 and NaCl content about 100 mmol/L. The presence of sodium diphosphate, [Na.sub.4][P.sub.2][O.sub.7], also increased stability of the MMT-[Na.sub.+] dispersions against coagulation by NaCl (21). However, adding a salt to clay suspension may enhance platelets aggregation and (upon drying) its crystals may be taken as impurity. Thus, washing the test specimen with demineralized water solves is essential.
Cadene et al. studied Wyoming bentonite, purified by dispersing it in demineralized water then ccntrifugation, stirring at 80[degrees]C for 12 h in water at pH = 5, and finally redispersed in 0.1 M NaCl solution for 12 h followed by filtration and repeated washing until total absence of [Cl.sup.-] ions (39). The authors also successfully dispersed the purified clay in NaCl solution (1 mmol/L) and measured individually selected exfoliated platelets dimensions using AFM. The reported data are: L = 320-400 nm, W [approximately equal to] 250 nm (hence L/W = 1.44 [+ or -] 0.16) and [t.sub.z] = 1.2 [+ or -] 0.1 nm. These values are not far from C-[Na.sup.+] dimensions in Tables 2 and 3.
Dynamic light/neutron scattering could be a useful technique for measuring the particle size, However, there are several potential problems related to sample preparation, clay and salt concentration, refractive index, platelet alignment in the flow direction and others, which may affect these measurements and influence the apparent size (15), (40).
Chemical Analysis of Clay Powder
The second aspect of these measurements is the chemical analysis of the original powder particles. In mineralogy, there are several test methods with complexity increasing upon the required degree of precision, including X-ray fluorescence instruments (XRF) or the electron probe microanalyzers (EPMA). The XRF precision for major elements (excepting hard to detect Al, Mg, and Na) is about 1% (41). The newer EPMA's (e.g., JXA-8530F) are highly sophisticated instruments, with magnification up to x50k, beam diameter 5-10 [micro]m and elemental sensitivity 0.005-0.01 wt% (42), (43). For the elemental analysis of clays the energy dispersive X-ray spectroscopy (EDS or EDX) used in conjunction with SEM has been used frequently (44), (45). Because of wide availability of this instrument, mainly the latter method was used in the VAMAS project.
The SEM-EDX, is a simplified version of EPMA, where the accelerated electrons from SEM gun eject the secondary electrons from the probed atoms. Compared to electron probe, SEM-EDX is easy to use. but for quantitative analysis, calibration of the instrument must be carried out (at the test voltage) using standards of known composition (similar to the test specimens). Furthermore, the results must be corrected for the background signal. SEM-EDX is about 10 times less sensitive than EPMA and suffers from several well-described artifacts (46).
As evident from data in Tables 4-6, the spread of values, collected in a single laboratory and the averages reported by different laboratories, is large. As the scatter is bigger for C-[Na.sup.+] than for ME-100, the local variability of C-[Na.sup.+] composition may be partially responsible. However, for ME-100 the spread of values reported by different laboratories is also uncomfortable, for example. for oxygen: O = 23-51 wt%; for sodium: Na = 1-5; for fluor: F = 5-29; hence, it significantly exceeds the reported instrument error and possible variability of clay composition. At the present stage, it is unclear whether this is a problem of calibration, uncorrected artifacts'. error in peak identification or all of these.
The last aspect of the clay analyses is the identification of impurities. Clarey et al. obtained patent for purification of natural clays involving 296 steps of grinding, dispersing, filtering, centrifuging, chemical treatment, hydrocy-cloning, etc. (21). As the document specifies, the resulting polymer-grade clay contains < 5 wt% ("preferably less than about 2% by weight") of impurities. The most difficult to remove are amorphous silicates, stacks of crystal-lographically welded platelets and the residual 0.3-2 wt% quartz particles with size > 300 nm. Additionally, purified clay such as C-[Na.sup.+] contains about 2 wt% of moisture and 7 wt% of loss on ignition, LOI (comprising organics, hygroscopic, and bound [H.sub.2]O, carbonic acid, etc.) (1).
The reported impurities of smectites include (22): oxides ([Fe.sub.2][O.sub.3], [SiO.sub.2] such as quartz, cristobalite and opal), silicates (albite. anorthite. biotite. feldspar, kaolinite. muscovite, orthoclasc, stilbite), sulfates (gypsum), carbonates (calcite, dolomite, siderite), phosphates (apatite), sulfites (pyrite). chlorides (sylvite, halite), etc. From Table 7, it is evident that in spite of complex purification most of these contaminants are present in the centrifuged sediments of C-[Na.sup.+] suspensions.
TABLE 7. Contaminants identified in centriliged sediments of C-[Na.sup.+]  Name Group Formula Montmorillonite Expandable [NA.sub.0.3][(AL, Mg).sub.2][Si.sub.4] [O.sub.10][(OH).sub.2][4H.sub.2]O Beidellite semecities [Na.sup.0.3][Al.sub.2] [(Si, Al).sub.4] [O.sub.10] [(OH).sub.2] * [2H.sub.2] Analcime Zeolite [Na.sub.2][Al.sub.2][Si.sub.4][O.sub.12] [2H.sub.2]O Quartz Oxides [SiO.sub.2] Cristobalile [SiO.sub.2] Rulile [TiO.sub.2] Anorthite Feldspar [CaAl.sub.2][Si.sub.2][O.sub.8] Microcline [KAlSi.sub.3][O.sub.8] Aragonite Carbonates [CaCO.sub.3] Vaterite [CaCO.sub.3] Dolomite CaMg[([CO.sub.3]).sub.2] Gypsum Sulphates [CaSO.sub.4] * [2H.sub.2]O Anhydrite [CaSO.sub.4] Alunite K[([Al.sub.3]([SO.sub.4])OH).sub.6]
The X-ray powder diffraction method was used for identification of C-[Na.sup.+] impurities. However, the drawback of this method is that if the data overlap, the structure determination is problematic. The Rietveld computer program method separates the overlapping peaks, thereby allowing for accurate determination of the structure. This method (also called Quantitative Phase Analysis, QPA) is quite powerful for determining the quantities of crystalline and amorphous components in multiphase mixtures (47), (48). Recently, Ufer et al. examined 36 contaminated bentonite samples from 16 locations, applying the newly modified Rietveld refined program for quantitative phase analysis, using corundum as an internal standard. The method is suitable for identification and quantification of clay impurities (49). However, the experimental error of mineral content by the QPA is still large, namely, 6 to 12% (50), (51).
As shown in Fig. 5, AFM is also helpful in analysis of clay impurities. As the lateral force microscopy (LFM) is sensitive to differences in chemistry of the surface its use with AFM offers an alternative approach.
This article summarizes the preliminary results from laboratories in seven countries working on development of the test methods for standardization of testing commercial clays destined for the PNC. Natural, semisynthetic, and synthetic clays were studied. The results show great polydispersity of these materials (polydispersity of shape, size, degree of dispersion, impurity content, and chemical composition). During the 80 years or so, the polymer industry learned to produce a great variety of specified grades of chemically identical polymers. Similarly, there is a great (but controlled) diversity of reinforcing materials for the classical composites. It is expected that in the future the nanofillers will have to be similarly controlled. Development of reliable procedures and test methods is the first step in that direction.
The reported results may be summarized as:
1. The clay platelets have irregular shape characterized by three orthogonal dimensions: length (L-the longest dimension), width (W-perpendicular to L), and thickness, [t.sub.z].
2. The number- or weight-averaged ratio: L/W = 1.4 [+ or -] 0.2.
3. The dry platelet thickness: [t.sub.z] = 1.09 [+ or -] 0.09 nm.
4. The effective platelet diameter: [D.sub.in] (nm) = 247 [+ or -] 37: 428 [+ or -] 97; and 658 [+ or -] 66 for C-[Na.sup.+], ME-100, and Topy-[Na.sup.+]. respectively.
5. The elemental composition of C-[Na.sup.+] and ME-100 (compare data in Tables 5 and 6) showed large errors within the same laboratory (from 9 to 52%), dependent on the atomic number and content. In spite of that, using the reported average values and calculating the molecular formulas (see the top four rows in Table 6) lead to consistent results. The scatter of chemical composition in part originates in the presence of impurities.
6. Only one laboratory analyzed the impurities. The method applied to C-[Na.sup.+] and ME-100 lead to identification of the main clay components (MMT and beidellite in C-[Na.sup.+] and fluoro hectorite in ME-100) and contamination of these two clays by 13 and 3 mineral, respectively.
Some laboratories encountered problems exfoliating clay (especially ME-100). and/or preventing reaggregation of clay platelets during drying on a PC-filter or mica surface. To prevent this, it is advisable to remove excess liquid and wash the sediment with demineralized water. The focus should stay on the development of better methods of clay exfoliation, possibly different ones for different type of clays. Furthermore, it is evident that human intervention is needed for identifying the well-defined exfoliated platelets, for the statistical size analysis. Thus, the original SEM/manual tracing method is worth pursuing, but so is the AFM (7), (38). In the latter case, the operator should select individual platelets or a computer program should reject the ones thicker than, for example, 1.9 nm. The diverse statistical methods of size analysis used in the Project also need scrutiny.
There is a significant discrepancy between expected precision of the SEM-EDX method for clay chemical analysis, namely, instrument specification of less than 0.2 % versus the errors in Tables 5 and 6 of > 9 %; the sources of this discrepancy should be identified. However, it is noteworthy that the literature data of natural clays composition show a similar scatter.
Identification of impurities in the natural C-[Na.sup.+] and semisynthetic ME-100 clays was performed in a single laboratory, thus it should be confirmed by other participants. It is also desirable that the test procedure quantifies the impurities. A large number of contaminants was expected for C-[Na.sup.+], but contamination of ME-100 by anthophyllite and gypsum was not. It is probable that small quantity of these were brought into the semi-synthetic clay with the key reaction ingredient, the natural talc.
Now-a-days purification of natural clays is not capable eliminating diverse impurities. Furthermore, their nature, shape, size, and chemistry depend on the source and their effects on performance are largely unknown. In consequence, better analytical methods should be devised to characterize these clays (or organoclays). The Rietveld method or the LFM with AFM offers good potential for analyzing the impurities.
The authors acknowledge the participation of INME-TRO's group, B. S. Archanjo, E. G. Gravina, and A. Yu. Kuznetsov, in experimental analyses and discussions. The UFCG coauthor is grateful to A. E. Zanini from UFBA for helpful analysis and comments. The CENAM coauthor is grateful to J. L. Cabrera, F. Rosas, E. Zapata, J. M. Juarez, and E. Ramirez, for the experimental analyses.
NOMENCLATURE P = [D.sub.in]/[t.sub.z] clay platelet effective aspect ratio AFM atomic force microscope CEC cation exchange capacity C-[Na.sup.+] Cloisite[R]-[Na.sup.+] d crystal spacing [d.sub.001] clay interlayer spacing [D.sub.in] average inscribed or effective diameter DLS dynamic light scattering EDS or EDX energy dispersive X-ray spectroscopy EPMA electron probe microanalyzer FEG-SEM SEM with the field-emission gun L, [L.sub.n], [L.sub.w] clay platelet length, its number and weight-averages LALS low angle light scattering LOI loss on ignition ME-100 Somasif ME-100 MIA micrograph image analysis MMT montmorillonite NPL National Physical Laboratory NRCC/IMI National Research Council Canada, Industrial Materials Institute PC polycarbonate PCS photon correlation spectroscopy PMCs polymer matrix composites PNC polymeric nanocomposites QPA Quantitative Phase Analysis SEM scanning electron microscope SEM-EDX energy-dispersive X-ray spectroscopy t ultrasonication time T temperature TEM transmission electron microscope Topy-[Na.sup.+] synthetic clay from Topy Industries TWA's Technical Working Areas TWA-33 Technical Work Area on Polymer Nanocomposites [t.sub.z] thickness of clay platelet VAMAS Versailles Project on Advanced Materials and Standards w clay content (wt%) w, [w.sub.n], [w.sub.w] clay platelet width, its number and weight-averages XRD X-ray diffraction XRF X-ray fluorescence
(1.) Southern Clay Products, Cloisite[R]-[Na.sup.+], Physical Properties Bulletin.
(2.) H. Tateyama, K. Tsunematsu, K. Kimura, H. Hirosue, K. Jinnai, and T. Furusawa, U.S. Patent 5204,078 (1993).
(3.) MSDS version 1.0.0 USA, Somasif ME 100. 25.10.2003.
(4.) L.A. Utracki, M. Sepehr, and E. Boccaleri, Polym. Adv. Technol., 18, 1 (2007).
(5.) Synthetic Mica, TOPY Ind. ltd., Aichi, JAPAN.
(6.) F. Perrin-Sarazin and M. Sepehr, Test Procedures for TWA-33, Project #1, Montreal, 11.06.2007.
(7.) S.M.L. Silva, P.E.R. Araujo, K.M. Ferreira, E.L. Canedo, L.H. Carvalho, and C.M.O. Raposo, Polym. Eng. Sri., 49, 1696 (2009).
(8.) L.A. Utracki, Clay-Containing Polymeric Nanocomposites, RAPRA, Shawbury, Shrewsbury, Shropshire, UK (2004).
(9.) I.F. Leite, A.P.S. Soares, L.H. Carvalho, C.M.O. Raposo, O.M.L. Malta, and S.M.L. Silva, J. Therm. Anal. Calorim., 100, 563 (2010).
(10.) J.L. Cabrera-Torres, F. Rosas-Gutierrez. E. Ramirez-Maldo-nado, J.M. Juarez-Garcia, and N. Gonzalez-Rojano, Report on the Determination of Shape Size and Size Distribution of Nano-filler Particles, Centro Nacional de Metrologia, Mexico, 28.08.2009.
(11.) L.H. Carvalho, C.A. Achete, B.S. Archanjo, A.Y. Kuznetsov, E.G. Gravina, and A.E. Zanini, Size and Size Distribution of Clays, 2-nd Annual Meeting of VAMAS TWA-33, Rome, 30.08.2009.
(12.) W. Broughton, Clay Platelets Dimensions, NPL Report, Teddington, UK. 20.08.2009.
(13.) Particle Size Analysis, Particle Size Distribution, Copyright 2010[C] HORIBA Instruments. Inc., and bibliography there.
(14.) M. Sepehr and F. Perrin-Sarazin, Determination of the Shape. Size and Size Distribution of Nano-filler Particles, NRCC/IMI Report, 21.05.2009.
(15.) B.G. Zanetti-Ramos. M.B. Fritzen-Garcia, C. Schweitzer de Oliveira, A.A. Pasa, V. Soldi, R. Borsali, and T.B. Creczyn-ski-Pasa. Mater. Sci. Eng.: C. 29, 638 (2009).
(16.) W. Broughton, Chemical Analysis of Clay by EDX, NPL Report. Teddington. UK. 16.09.2009.
(17.) G. Camino, Report on TWA-PNC Project #1, VAMAS TWA-33 General Meeting. 30.08.2009 - Rome. Italy.
(18.) M. Sepehr, F. Perrin-Sarazin, and L.A. Utracki, PA Nanocomposites Structural Parameters/Mechanical Performance, NRCC/IMI, the 9th PNC-Tech Meeting, 12.06.2007.
(19.) V.A. Drits. Clay Minerals, 38, 403 (2003).
(20.) G. Lagaly and S. Zismer, Adv. Colloid Interface Sci., 100-102, 105 (2003).
(21.) M. Clarey, J. Edwards, S.J. Tsipursky, G.W. Beall, and D.D. Eisenhour, U.S. Patent 6050,509 (2000).
(22.) P. Schick, U.S. Patent 3865,240 (1975).
(23.) D.D. Bilanovic. S.A. Spigarelli, and T.J. Kroeger, EURO-SOIL, Freiburg, Germany, 04 - 12.09.2004.
(24.) L.J. Arroyo. H. Li, B.J. Teppen, and S.A. Boyd. Clays Clay Min., 53, 511 (2005).
(25.) S.S. Araujo. P.E.R. Araujo. C.M.O. Raposo. L.H. Carvalho, and S.M.L. Silva, Anais do XVI Congresso Brasileiro de Engenharia e Ciencias dos Materials - CBECIMat. 2006. Foz do Iguacu.
(26.) S.S. Araujo, P.E.R. Araujo. S.M.L. Silva, C.M.O. Raposo, and L.H. Carvalho, Anais do XVII Congresso Brasileiro de Engenharia Quimica COBEQ, 2008. Recife - PE.
(27.) J.W. Earley, B.B. Osthaus, and I.H. Milne, Amer. Mineralogist, 38. 707 (1953).
(28.) H.E. Roberson, A.H. Weir, and R.D. Woods, Clays Clay Miner., 16. 239 (1968).
(29.) M. Okamoto, Personal Communication on the Method of ME-100. December 2009.
(30.) F.R. Huege, US Patent 3939,249 (1976).
(31.) G.W. Beall, S. Tsipursky, A. Sorokin, and A. Goldman, U.S. Patent 5552,469(1996).
(32.) T. Lan and E.K. Westphal, U.S. Patent 6251,980 (2001).
(33.) J. Temuujin, K. Okada, T.S. Jadambaa, K.J.D. MacKenzie, and J.J. Amarsanaa, European Ceramic Soc, 23, 1277 (2003).
(34.) L. Perez-Maqueda, F. Franco, M.A. Aviles, J. Payato, and J.L. Perez-Rodriguez, Clays Clay Minerals, 51, 701 (2003).
(35.) A.L. Poli. T. Batista, C.C. Schmitt. F. Gessner, and M.G. Neumann. J. Colloid Interface Sci., 325, 386 (2008).
(36.) E.B. Flint and K.S. Suslick, Science, 253, 1397 (1991).
(37.) T. Missana and A. Adell, J. Coll. Interface Sci., 230, 150 (2000).
(38.) E. Tombacz and M. Szekeres, Appl. Clay Sci., 27, 75 (2004).
(39.) A. Cadene, S. Durand-Vidal, P. Turq, and J. Brendle, J. Colloid Interface Sci., 285. 719 (2005).
(40.) D.S. Jayasuriya, N. Tcheurekdjian, C.F. Wu, S.H. Chen, and P. Thiyagarajan, J. Appl. Cryst., 21, 843 (1988).
(41.) P.J. Potts. P.C. Webb, and J.S. Watson. X-Ray Spectrometry, 13. 2 (1984).
(42.) S.J.B. Reed. Electron Microprobe Analysis and Scanning Electron Microscopy in Geology, 2nd ed., Cambridge University Press, Cambridge, UK (2005).
(43.) J. Goldstein, D.E. Newbury. D.C. Joy. C.E. Lyman. P. Ech-lin, E. Lifshin, L. Sawyer, and J. Michael, Scanning Electron Microscopy and X-Ray Microanalysis. 3rd ed., Kluwer Academic/Plenum Publishers, New York (2003).
(44.) A. Wiewiora, P. Giresse, S. Petit, and A. Wilamowski, Clays Clay Minerals, 49. 540 (2001).
(45.) B. Schoene and S.A. Bowring, Contrib. Mineral, Petrol., 151, 615 (2006).
(46.) ASTM E1508 - 98(2008) Standard Guide for Quantitative Analysis by Energy-Dispersive Spectroscopy.
(47.) B. Peplinski, R. Kleeberg, J. Bergmann, and J. Wenzel, Mater. Sci. Forum, 443-444, 45 (2004).
(48.) D.L. Bish and S.A. Howard, J. Appl. Cryst., 21, 86 (1988).
(49.) K. Ufer, H. Stanjek, G. Roth. R. Dohrmann, R. Kleeberg, and S. Kaufhold, Clays Clay Minerals, 56, 272 (2008).
(50.) T. Monecke, S. Kohler, R. Kleeberg, P.M. Herzig. and J. B. Gemmell, Can. Mineral., 39, 1617 (2001).
(51.) M.E. Alves and O. Omotoso, Soil Sci. Soc. Am. J., 73, 2191 (2009).
Correspondence to: Dr. Leszek Ulracki; e-mail: email@example.com
Published online in Wiley Online Library (wileyonlinelibrary.com).
[C] 2011 Society of Plastics Engineers
Leszek A. Utracki, (1) Bill Broughton, (2) Norma Gonzalez-Rojano, (3) Laura Hecker de Carvalho, (4) Carlos A. Achete (5)
(1) National Research Council Canada, Industrial Materials Institute, 75 de Mortagne, Boucherville, QC, Canada J4B 6Y4
(2) Bio, Polymeric and Composite Materials, Industry and Innovation Division, National Physical Laboratory, Materials Division, Teddington, TW11 OLW, UK
(3) Centro Nacional de Metrologia (CENAM), Division de Materiales Organicos, km 4,5 carretera a Los Cues, El Marques, Queretaro, QT, C.P 76 241, Mexico
(4) Universidade Federal de Campina Grande, Centro de Ciencias e Tecnologia, Av. Aprigio Velososo 882, Bodocongo, Campina Grande, Brazil
(5) Instituto Nacional de Metrologia, Normalizacao e Qualidade Industrial - INMETRO, Divisao de Metrologia de Materials, Av. N.S. das Gragas 50, Duque de Caxias, RJ, Brasil
DOI 10.1002/pen. 21807
|Printer friendly Cite/link Email Feedback|
|Author:||Utracki, Leszek A.; Broughton, Bill; Gonzalez-Rojano, Norma; Carvalho, Laura Hecker de; Achete, Carl|
|Publication:||Polymer Engineering and Science|
|Date:||Mar 1, 2011|
|Previous Article:||Reinforcing and toughening of polypropylene with self-assembled low molar Mass additives.|
|Next Article:||Analysis of tensile test results for poly(acrylonitrile-butadiene-styrene) based on Weibull distribution.|