Wavelength dependent haze of transparent glass-particle filled poly(methyl methacrylate) composites

: Glass particles as filler were incorporated in a poly(methyl methacrylate) matrix. The refractive indexes of both materials match at a wavelength of about 400 nm. The effect of particle volume fraction on the light transmittance and light scattering (haze) in dependence of the refractive index difference was studied. The curve shape of the haze in dependence of the wavelength is comparable to that of the refractive index difference, but the base line of the haze increases with the filling grade. This indicates that there are other scattering or absorbing mechanisms, like defects in the filler binding.


Introduction
Particle-filled Polymers are normally non-transparent. By the use of a transparent polymer as a matrix and an optical glass with a matching refractive index n as filler particles, a transparency could be achieved for at least one temperature and wavelength of light. These transparent composites are expected to have, compared to the bulk material, increased mechanical properties (higher stiffness, increased strengths by the use of fibers) a lower expansion coefficient an increased thermal conductivity. In the case, when the particle size d is one order of magnitude greater than the wavelength of light λ in the visible range (d >> λ) the light scattering theories of Rayleigh and Mie cannot be applied. [1] First experiments on this subject have been done by Breuer and Grzesitza [2]. They analyzed a mixture between two polymers to adapt the refractive index to that of the used glass fibers. Specimens show a wavelength-(because of the different dispersion) and temperature-dependent extinctioncurve of the indecent light. Significant work on this subject has been carried out in Japan by the group of Kagawa. They studied the influence on light transmittance, mechanical and thermal properties of many parameters, such as refractive index difference, particle size, particle surface area and filler content. [3][4][5][6][7][8]. The investigated materials are an epoxy resin as a matrix and different optical glasses as filler particles. The light transmittance clearly shows a dependence of the refractive index difference because of the different dispersions of the materials. Nevertheless, the light transmittance of the compound is always less than that of the pure matrix. Weaver and Stoffer [9] investigated glass-fiber reinforced PMMA. They used optical glass as fiber material and studied the optical properties in dependence of the temperature among others. Experiments show that the temperature dependence of the polymer changes with the filling grade due to the reduced thermal expansion of the compound.
To archive a transparent composite, a matching refractive index of filler particles and polymer matrix is necessary. The refractive index of a specific material changes with the wavelength of the incident light (dispersion) and with the temperature of the material. These two dependencies are not the same for commercial available polymers and glasses. Therefore 2 Experimental Procedure

Materials
As matrix, the polymer poly(methyl methacrylate) (PMMA, Plexiglas® 7N by Evonik Industries AG) and as filler the glass "N-PK52A" (Schott AG) was chosen. This glass was selected to obtain a matching refractive index with the polymer matrix for at least one wavelength in the visible range at room temperature.  [-] 55,82 (2) 81.52 (1) Table 1: properties of the materials used (1) according to supplier (2) in-house measurement

Milling of glass and characterization of particles
The bulk glass was crushed with a mortar and then thieved by a sieving machine with two laboratory sieves and a mesh width between 63 and 180 µm. Afterwards, the glass was washed with acetone to remove small particles. The glass particles were characterized by SEM-images (SEM Ultra Plus type, supplier: Zeiss), the volumetric and numeric particle-distribution was determined by the measuring instrument "Morphology" of Malvern Instruments. For each measurement, at least 42000 particles were photographed and analyzed.

Fabrication of composite
To incorporate glass particles into the polymer matrix the micro-compounder "HAAKE MiniLab" by Thermo Fisher Scientific Inc., with two conical screws was used. The temperature was set to 210°C. To distribute the particles equally in the melt a circle mode was held up for at least 3 minutes. Subsequently specimens were pressed with a thickness about 250 µm and a diameter of 20 mm. The particle volume fraction was controlled by thermal gravimetric analysis (TGA) measurements.

Refractive index
The refractive Index of the polymer was determined with the refractometer

Haze
The Haze of the specimens was analyzed with an UV/VIS spectrometer (Lambda 18 by Perkin Elmer Inc.). Therefore, the transmission and the light scattering of the specimens were investigated at different wavelengths. Figure 1 shows the fundamental set up for these measurements. For transmission measurements, all the incoming light is collected by the sphere and so measured by the detector (a). If the beam trap is installed at the opposite position to the opening of the sphere, only the scattered light is measured by the detector.
The undistracted light beam is completely absorbed by the beam trap (b).

Figure 1: UV/VIS spectrometer with the setup for transmission measurements (a) and measurements of the scattered light (b)
The Haze was calculated using the following equation according to [10].
Haze in % = � To determine the haze only caused by the particles and not by the surface of the specimen, the haze of a non-filled specimen was subtracted by the haze of the filled specimen.

Particles
The numeric particle distribution is shown in Figure 2. The smallest particles observed have a diameter of 2 µm. The volumetric particle distribution of washed particles is characterized as following: d 10,3 = 116.3, d 50,3 = 178.7, d 90,3 = 246.7. In Figure 2 can be seen that there is, despite of sieving and washing, a considerable amount of small particles (d < 63 µm).
Nevertheless, the volumetric amount of these particles is quite small so that they should have a relatively small influence on the light scattering in the compound. Figure 3 shows the appearance of the glass particles, observed by scanning electron microscopy. According to the production process, they have an irregular shape and vary in size.   (2)

Figure 2: numeric distribution of the unwashed and washed particles
The curve progression of the glass was calculated with the Sellmeier Equation (3) with constants according to the data sheet.

Figure 5: refractive index of PMMA Plexiglas 7N and glass N-PK52A at 23°C
Observations of the refractive index difference by the oblique illumination method at different wavelengths nearly confirm these measurements, Figure 6. At a wavelength of 400 nm the particle is nearly invisible. At longer wavelengths a shadow at the side of the particle facing the less illuminated side of the picture can be clearly seen. This indicates that the refractive index of the particle is higher than that one of the matrix.

Conclusion
The effect of the refractive index difference between a polymer matrix and glass particles as filler on the haze has been studied. The haze shows a minimum at the wavelength with matching refractive indexes and the change of haze increases with higher filling grade.
However, the minimum haze is about 10 % and 5 % for the filling grades of 10 vol.-% and 15 vol.-%. Further work will deal with the high haze for higher filling grades and set up a relation between refractive index difference, haze and clear view through the specimens.

Acknowledgment
The authors gratefully acknowledge the German Research Foundation (DFG) for the funding of the work, the Institute of Polymer Materials for the use of their micro-compounder, and the industrial partner Evonik Industries AG for providing material.