Analysis of the Powder Composite Process by a Mechanical Method t

Recently, composite particles covered with different kinds of fine particles have been fabricated using various mechanical methods in a dry phase. However, the quantitative analysis of the composite process has not yet been conducted systematically. In this paper, a mechanical processing method called MECHANOFUSION was used to investigate the composite process of glass beads and titanium dioxide fine particles. As a first trial, the ratio of fine particles fixed onto core particles and the BET specific surface area of the processed particles were measured as a function of the processing time. Consequently, the composite process was described as the following two steps: The first is the adhering step whereby fine particles adhere to the core particles' surface, and the second is the compacting step of the fine particle layers. Furthermore, it was found that the BET specific surface area of the processed powder was correlated with the energy consumption per weight of material in the apparatus.


Introduction
Recently, attempts to fix fine particles onto the surface of particles to produce the so-called composite particles is being actively pursued in the field of powder technology.
A number of processing methods have been proposedll but the dry type mechanical method 2 l proposed by Koishi et a!. is known to be simple and to possess high applicability.Motivated by this success, a number of methods using fine grinding machines 3 l have been developed in succession 2 >.Also there are a number of reports 3 • 5 > which claim that the production of composite particles in a variety of combinations including plastics, metals, ceramics and such has become possible and numerous studies on the characterization 6 • 7 l and application 8 • 10 l of the obtained particles are documented.
Thus, studies in these areas are being actively pursued but presently, not many studies of the composite process are being carried out by actual machines 11 • 12 l.It is believed that the main reason t This report was originally printed in ].Soc. Powder Technology, japan, 29, 434-439 (1992) in Japanese, before being translated into English with the permission of the editorial committee of the Soc.Powder Technology, Japan.
KONA No.ll (1993) for this lies in the difficulty to perform a quantitative analysis of the obtained composite particles.In fact, much of the analysis of characterization and composite process phenomenon is conducted mainly by electron microscopes.
In this present study, one of the actual machines, the Mechanofusion system 4 > (manufactured by Hosokawa Micron Corp.) was used to observe the characterization of composite particles from their structure and an attempt was made to analyze the observations using a simple and quantitative method.Then on the basis of the results, the particle composite process was analyzed.Furthermore, the relation between the operation conditions of the apparatus and the composite process was studied.

Experiment
The outline of the experimental apparatus is shown in Fig. 1.The main part consists of a rotating chamber and a semi-cylindrical arm head fixed with a certain clearance against the inner surface 'of the chamber.Powder loaded into the chamber is compressed onto the inner surface by centrifugal force and receives the complicated forces such as compression, shearing and rolling between the arm head and the inner surface of the chamber.The powder subjected to this action is scraped off from the inner surface by the scraper and again subjected to the above action.It is thought that the composite particles are obtained by repeating this action during processing.In the experiment, AM-20F shown in Table 1 was used but the other three types of facili- ties were also used for the data mentioned later.
In setting the experimental conditions, the combination of particles and the mixing ratio are defined from the aspect of the materials while for the operation conditions, rotation speed of the apparatus, atmosphere, temperature and such are determined.Here, for the purpose of analyzing the composite process, the ratio of the core particles diameter to that of the fine particles was set 2 l at a large value which is said to be effective to fabricate composite particles.Therefore, the powder specimens men-tioned in Table 2 were used.
The glass beads used in the basic experiment are spherical test specimens of GMB-20 powders with uniform particle diameters.PMMA is of spherical shape while silicon sand is of irregular shape.For the fine particles which were fixed onto the surface of the core particles, rutile type titanium dioxide (produced by Teika Co., MT-150W) was used.The core particle size was measured by the laser analysis method (Microtrac) and the average particle size was expressed as 50o/o diameter of the volume basis.The average diameter of the titanium dioxide is catalog value.The weight ratio of titanium dioxide to the glass beads was set at 0.03.As for the rotation speed, 3 conditions were set to prevent crushing of the glass beads.The present experiments were all conducted in normal atmosphere and no external heating or cooling of the apparatus was provided.
In the experiment, the core particles and the fine particles were weighed for the specified ratio, and a total of 500 grams of the particles was loaded in the chamber and rotated at the set speed for the specific time.The motor power during operation was measured and by subtracting the value at idling, a value corresponding to the power acting on the powder was obtained.The temperature of the chamber during operation was obtained through a thermocouple embedded in the arm head.In the case of processing glass beads and titanium dioxide, the chamber temperature did not exceed 361K.
First, to observe the fixed ratio (R) of titanium dioxide onto the glass beads with the lapse of processing time, a simplified method using the apparatus shown in Fig. 2 was employed.After mixing 5 grams of the processed powder in methyl alcohol for 1 minute, the powder was dispersed for 1 minute by ultrasonic waves (300 W).Then the powder was passed through a micro-sieve (mesh 10 ± 1 11m, Tsutsui Rikagaku Kikai Co.) by sound wave oscillation for 90 minutes while adding methyl alcohol.The residue left on the sieve was dried for 24 hours Specific surface area (m 2 ;kg) 0.062 X 10 3 "" 0.42 X 10 3 .. 0.59 X 10 3 ' 97.9 X 10 3 " *BET ••calculated

Methanol
Micro sieve Speaker _ _ _ _ c:::.:::======::::::JFig. 2 Experimental apparatus for measuring the ratio of fine particles fixed onto the surface of core particles in a constant temperature bath (393K), the titanium dioxide residue was weighed, and from this, R was calculated.The validity of this method was confirmed by making use of the methyl alcohol into which 0.15 gram of titanium dioxide and 4.85 gram of glass beads are dispersed.In this case, R was -0.015 under the above processing condition.It proves the error is little in this method.Furthermore, R will change with the dispersion conditions of the processed powder.However, from the viewpoint of discussing the composite process, the experiments were conducted under the above mentioned constant conditions.

Experimental results
Fig. 3 shows the relationship between the fixed ratio R onto the surface of glass beads and the processing time t.Here, when we look at the effect of rotation speed, the spread of data is observed at 500 rpm, but R increases with processing time and becomes constant at about 0.9 in 5 minutes or more.With rotation speed exceeding 750, R reaches 1.0 KONA No.ll (1993) in 5 minutes.Fig. 4 shows SEM photos of composite powder after processing for 120 minutes at 500 and 1000 rpm and for both, it can be observed that titanium dioxide has fixed onto the surface of the glass beads to form composite particles.
From the above, the relationship between the ratio Rand the processing time tis described by the following two steps: The first one is a step in which fine particles adhere onto the core particle.The second one is a step in which R becomes constant.The former step is seen earlier as rotation speed increases.
Next, in an attempt to observe how the structure of the particles changes in the above composite process, the BET specific surface area was measured.An automatic specific surface measuring apparatus Model 2200-01 manufactured by Shimadzu Corp. was used and N 2 was used as the absorption gas.To obtain information limited to composite particles, those of R = 1, namely samples processed at 750 rpm and 1000 rpm were used.
Fig. 5 shows the relationship between the specific surface area Sw of the processed powder and the processing time t.Here, Sw at t = 0 was obtained by calculating it from the values in Table 2. From the Fig. 5, it can be seen that Sw decreases, tracing an S curve with the lapse of time.As for the effect of rotation speed, it is seen that the Sw value is smaller at 1000 rpm than that at 750 rpm.Looking at Fig. 4 (c) to study the cause for the decrease of Sw, it can be observed that titanium dioxide is present in a compacted condition.Consequently, it is thought that the reduction in specific surface area is caused by the compaction of the titanium dioxide layer onto the glass bead surface.From the above, in the area where R becomes constant, the specific surface area decreases as the processing time becomes longer, and the surface layer of core particle tends to be compacted.A similar phenomenon is also observed when the core particles are substituted with PMMA.For the series shown in Fig. 6, it has not been firmly confirmed that the fixed ratio of fine particles onto core particles is 1.However, a trend similar to that of Fig. 5 is seen.Fig. 7 shows electron microscope photos of processed particles with processing times of 5 minutes and 480 minutes.It is observed that the compacting density of the titanium dioxode layer increases with the reduction of the specific surface.
From the above example, it is thought that the phenomenon explained in Fig. 5 is quite general.Also since the specific surface in Fig. 6 becomes roughly constant with an extended processing time, it is assumed that a similar trend will be seen in the case of Fig. 5 when processing is conducted for a long time.
232 Here, I is considered to be the step where fine particles adhere onto the surface of the core particle and as a result, R increases with the lapse of time while contrarily Sw tends to decrease.On the other hand, in II, the adhering of fine particles onto the core particle surface ends and only Sw decreases which is considered to be a compacting process of the fine particle layer on the core particle surface.As already known, this model, was analyzed as bulk material.Therefore, it is thought that a study of its distribution is also necessary in the future.
For practical purposes, it is important to know the relationship between the composite process and the operation conditions of the apparatus.Here we will focus on the case where all fine particles fix onto the core particle (R = 1) which is an important area from the viewpoint of practical use.Looking at Fig. 5, it can be seen that the specific surface area Sw becomes smaller as the processing time t increases and also as rotation speed is increased.This implies that Sw becomes smaller as the energy supplied to the powder becomes higher.As a trial, if the x axis represents the energy E applied to a powder unit mass, we obtain Fig. 9 and it shows that the data is expressed by one curve independent from rotation speed.Therefore, in the area where R = 1, the setting of the operation conditions is made by considering E. Fig. 10, shows data which were re-arranged by the present method.In the experiments, the 4 types of different sizes, shown in Table 1 were used.The properties of silica used as the core particles are shown in Table 2. Here, the fixing ratio R of titanium dioxide onto silica surface was not measured.However, it is interesting to note that  the data can be arranged on one curve without referring to the size of the apparatus and its operation conditions.This implies that the scale-up of the apparatus can be done by the relation shown in Fig. 10.It is necessary to have further studies hereafter on this point, including the measurement of R values.

Conclusions
Using an actual machine and expressing the particle composite process by the fixing ratio of fine particles onto core particles and by the specific surface area of the processed powder quantitatively, the following conclusions were obtained.(1) The particle composite process obtained with Mechanofusion apparatus was described by the following two steps: the step where fine particle adheres onto the core particle surface and the step where the fine particles compact onto the core particle surface.
(2) The composite process was correlated with the energy applied to a powder unit mass.

1. 2 .Fig. 3
Fig. 3 Change of the ratio of fine particles fixed onto the surface of core particles with processing time (glass beads -Ti0 2 )

Fig. 5
Fig. 5 Change of BET surface area of composite particles with processing time (glass beads -Ti0 2 )

Fig. 9
Fig. 9 Relationship between BET surface area of composite particles and energy consumption (glass beads-Ti0 2 )

Fig. 10
Fig.10Relationship between BET surface area and energy consumption of four kinds of apparatus indicated in Table1(silica sand-Ti0 2 )

Table 1
Specifications of experimental apparatus

Table 2 .
Properties of powder materials used