Halloysite Nanocomposites

Abstract

[evdate 2009-05-01] The Hong Kong University of Science and Technology Department of Chemical and Biomolecular Engineering Final Report of President's Cup Project Title: Halloysite Nanocomposites

General Information Year of Name Student ID School Department study Chan Hin Chung School of Engineering CBME 3 Ho Sze Ting School of Engineering CBME 3 Lee Chun Ning School of Engineering CBME 3 Lui Yat Chuen School of Engineering CBME 3

Supervisor: Professor Chi Ming CHAN, Dr. Kai Mo NG Endorsed by:

Prof. Chi Ming CHAN Department of Chemical and Biomolecular Engineering Date: 1st April, 2009

1

E-mail address Phone number ! The Hong Kong University of Science and Technology Department of Chemical and Biomolecular Engineering Final Report of President's Cup Project Title: Halloysite Nanocomposites

Amendment Sheet This document aimed at amending the two minor typos occurred in the hardcopies of the President's Cup Final Report. The two typos do not affect the meanings of the content, but it is strongly recommended to take a look at this sheet first, which can ensure a more smooth reading of the Final Report. You may also refer to the Final Report in pdf format, which is the error-free finalized version.

Amendment 1 Page 16: 'For ball-milled samples, the diameters of the agglomerates were determined from SEM photos, Figure 3.6(d) and 3.6(e).' to 'For ball-milled samples, the diameters of the agglomerates were determined from SEM photos, Figure 3.7(d) and 3.7(e).'

Amendment 2 Page 17: 'Therefore, combining Table 3.3, 3.4, 4.1 and 4.2, the impact strength ... ' to 'Therefore, combining Table 3.2, 4.1 and 4.2, the impact strength ... ' Thank you very much and sorry for any inconvenience caused. Best regards, Group

Abstract The effect of different micro-scale and nano-scale fillers to toughen polymer is undeniable, and the toughening mechanisms are well-known and widely studied. Yet, there is a rarely considered area, which is the effect of the size of these filler on the degree of toughening. This project reveals stunning findings and results on this area, which is a big step in the science of fillers. The utilization of nanotechnology is booming. Scientists are being inspired to widen its areas of application every day, all around the world. One such application is the incorporation of halloysite nanotube (HNT) into epoxy, which has shown that its mechanical properties have been improved [1]. Inspired by the great achievement, we synthesized polystyrene/halloysite nanocomposite with halloysite particles of different agglomerate sizes. It was found that the impact strengths of the nanocomposites were improved, and the extent of improvement was dependant on the sizes of incorporated agglomerates. The preparation of patent filing of our experimental procedure is now in progress.

2

Table of Contents 1. Introduction 4 2. Experimental 5 2.1 Treatment of raw halloysite 5 2.1.1 Preparation of micro-scale halloysite agglomerates 5 2.1.2 Preparation of nano-scale halloysite particles (HNT) 5 2.2 Preparation of nanocomposites - Twin Screw Extrusion 6 2.3 Materials' characterization 6 2.4 Mechanical tests 6 2.4.1 Charpy Impact Test. 6 2.4.2 Injection Molding 7 3. Results 8 3.1 Morphology of halloysite 8 3.2 Mechanical properties 8 3.3 Toughening mechanism 12 4. Discussion 16 4.1 Relationship between agglomerate size and impact strength 16 4.2 Relationship between number of agglomerates and impact strength 18 4.3 Patent filing 19 5. Conclusion 19 6. References 20

3

1. Introduction Halloysite is a clay material with the empirical formula AI2Si205(OH)4 consisting of numerous numbers of nanotubes (Halloysite nanotube, HNT) [1]. Its main constituents are aluminium, silicon, and hydrogen, while HNTs are tubular halloysite molecules with diameter generally smaller than 100nm.

The addition of HNT, which is inexpensive, has been shown to significantly increase the fracture toughness of the epoxy. Increases in fracture toughness are mainly due to mechanisms such as crack bridging, crack deflection and plastic deformation of the epoxy around the halloysite particle clusters [2]. It has effectively improved the strength of the nanocomposite by almost four times [1]. Based on a similar argument, it is possible that the similar phenomenon may be observed in polystyrene (PS), which is also a brittle polymer.

We blended halloysite agglomerates of different sizes with neat PS. These different agglomerate sizes could be further classified under micro-scale and nano-scale. It was shown that these nanocomposites demonstrated different levels of strength. Improvements in impact strength were especially remarked with nano-scale agglomerates, which achieved as high as three times that of neat PS. Apart from this, there is also a very clear dependence of impact strength on agglomerates' sizes.

4

2. Experimental 2. 1 Treatment of raw halloysite 2. 1.1 Preparation of micro-scale halloysite agglomerates The nanocomposites were prepared by mixing 95wt% polystyrene (Styron* 685D 26 7) and 5wt% halloysite (MM). Agglomerates of naturally occurring halloysite are several hundred microns in size. Nested sieves with openings of 180μ, 125μ, 90μ, 63μ and 45μ were used to separate agglomerates with similar sizes. In brief, sieves with decreasing sizes of openings were nested from the upper position to the lower position, and at the bottom a tray was situated to collect the residue. Raw halloysite was dried in oven at 80°C for 14 hours, which was then uniformly added onto the top sieve, and separated mechanically by the shaker for 15 minutes. Agglomerates of different size ranges were then collected. Conclusively, the ranges of the particles' diameters were <45 μ , 45μ -63μ , 63μ-90μ , 90μ-125μ and 125 μ -180μ .

2. 1.2 Preparation of nano-scale halloysite particles (HNT) As the cohesive forces between HNTs are strong, they have a tendency to form stable agglomerates, where nested sieves and mechanical shaker cannot provide enough energy to further break down to give HNTs. In order to obtain HNTs from raw halloysite, a ball mill (Machine Number: 2384C, made by the Pascali Engineering), which grinded raw halloysite agglomerates into much finer particles was used. Halloysite particles within the size range 63μ -90μ were first dried in oven at 80°C for 14 hours. Dried halloysite was then transferred into a ball mill. Grinding were performed for 2 hours, 4 hours, 6

5

hours and 14 hours to give HNTs of different sizes, they were then collected.

2.2 Preparation of nanocomposites - Twin Screw Extrusion PS was mixed with either micro-scale or nano-scale halloysite particles in a twin screw extruder (Collin Teach-line® ZK25T).

Dry PS and halloysite were uniformly mixed in a plastic bottle by manual shaking. The mixture was poured into a twin-screw extruder. The condition of the extruder was set as: temperatures from hopper to die were set at 150°C, 240°C, 250°C, and 240°C, respectively [3]; screw speed at 80rpm and the feed rate at 140rpm. The product was in the form of thin, cylindrical solid tubes, which would then be pelletized by a pelletizing machine. To make sure the halloysite particles were mixed more evenly with PS, the PS/HNT nanocomposite pellets were fed into the twin screw extruder again. Therefore, the final pellet form products could be described as 'double-twin-screw-extruded' .

2.3 Materials' characterization The morphologies of the halloysite and PS/halloysite nanocomposites were examined using (JEOL JSM-6390), Scanning Electron Microscopy (SEM).

2.4 Mechanical tests

2.4. 1 Charpy Impact Test To find out the toughness of the nanocomposites, Charpy impact test was

6

performed following the ASTM-D256 method, by the Cantilever Beam Impact machine. The specimens had the same size of 70.0 x 10.0 x 3.0 mm3 , which were tested using an impact tester (Tinium Olsen 92T), and the hammer was pendulum-type. The condition of the impact test was set as: temperature at 23°C, relative humidity at 50% and pressure at 1 atm. A sharp pre-crack was produced by notching using a CSI Automatic notcher (CS-93M), the table feed rate and the cutter speed were 100mm/min and 92m/min, respectively. To investigate the crack initiation process, the fracture surface was examined with SEM.

2.4.2 Injection Molding This experiment was performed to prepare the specimens of impact tests. The Morgan-Press injection machine was used. The condition was set as: barrel temperature at 200°C, nozzle temperature at 210°C, compressed air at 160psi, injection pressure at 4000psi, internal molding pressure at 9tons, and relative humidity at 50%.

Sleving halioysite Ball-milling halloysite Porystyrene Twin Screw Extrusion PS/halloysite Composite Injecton molding Charpy ImpactTest Figure 2.1 Flow chart of the experiments

7 SEM 3. Results 3. 1 Morphology of halloysite Figure 3.1 shows the SEM micrographs of halloysite. It can be seen that this material contains some straight tubes which are HNTs, and the majority is halloysite particles, which come from the agglomeration of HNTs. (a) (b) Figure 3.1 SEM of halloysite agglomerate sizes of the range (a) 125I-1m-180l-lm, and (b) <45μm.

3.2 Mechanical properties For the nanocomposites of each agglomerate size range, 10 samples were used for each test. Table 3.1 summarizes the average impact strengths of all the samples. To begin with, we present the raw data obtained, as shown in Table 3.1 and Figure 3.2. The average impact strength of neat PS was found to be 18.4J/m, while that of nanocomposite filled with halloysite ball-milled for 2 hours (57.4J/m) is

8

outstandingly three times that of neat PS. To our surprise, the impact strengths of them follow an impressive trend, which increases as the agglomerate size decreases, until it reaches the peak value (57.4J/m) given by halloysite ball-milled for 2 hours, and then the energy drops beyond this optimum agglomerate size. Starting from the 3-hour sample and up to the 6-hour sample, the impact strength is consistently maintained at around 30J/m, which is 1.63 times of that of neat PS. After 3 hours of ball-milling, the agglomerates could not be grinded further into finer particles by the ball mill. In other words, the maximum capability of the ball mill was reached.

Table 3.1 Impact strength of halloysite composite with: (a) sieved halloysite and neat PS, and (b) ball-milled halloysite. Impact strength (J/m) Size range (μm) I Neat PS

63-90 45-63 Neat PS <45 1st sample 7.51 12.0 11.5 21.1 2nd sample 7.87 12.0 12.3 22.0 3rd sample 8.00 12.0 13.3 23.2 4th sample 8.23 12.1 19.8 26.8 5th sample 8.26 12.6 20.5 29.7 6th sample 8.68 13.7 20.6 32.4 7th sample 9.56 15.2 20.6 32.5 8th sample 9.85 15.6 21.6 37.7 9th sample 10.5 16.8 21.9 39.6 10th sample 14.7 17.0 22.4 55.6 (a)

9

Impact strength (J/m) Duration of ball-milling (hour) 1 2 3 4 6 1st sample 16.3 44.8 16.7 24.3 20.7 2nd sample 20.0 46.1 23.0 26.0 21.3 3rd sample 23.5 48.3 24.7 26.5 22.6 4th sample 25.6 50.8 25.5 28.4 25.5 5th sample 27.1 51.3 29.3 28.4 26.5 6th sample 27.8 53.5 32.2 28.6 29.1 7th sample 28.4 54.8 32.2 29.2 30.7 8th sample 34.5 61.5 32.6 31.9 31.9 9th sample 41.6 68.8 34.1 33.3 39.6 10th sample 46.4 94.4 40.0 47.1 43.5 (b) 60 50 30 Pi E 20 (a)

10

< Neat PS 1 2 3 4 6 14 Duration of ball-milling (hour). (b) Figure 3.2 Impact strength of halloysite composite with: (a) sieved halloysite and neat PS, and (b) ball-milled halloysite One point to note is that nanocomposites with exceptionally high impact strengths were found, such as 55.6J/m from <45μm and 94.3J/m from ball-milled 2 hours' sample, which are 3 times and 5 times of that of neat PS, respectively. These results could have been coming from the optimal agglomerates which were able to give the best improvement out of their belonged ranges.

11

Table 3.2 Average impact strength of halloysite composite with: (a) sieved halloysite and neat PS, and (b) ball-milled halloysite Diameter range of 63-90 45-63 <45 agglomerates (um) Impact strength 9.32 13.9 32.0 (J/m) (a) Duration of 1 hr 2hr ball-milling (hour) 3hr 4hr 6hr 14hr Impact strength 29.1 57.4 (J/m) 29.0 30.4 29.1 30.6 (b)

On the other hand, incorporating halloysite agglomerates with diameter larger than 45IJm has negative effects on PS. The average impact strengths of 451Jm-631Jm and 631Jm-90lJm samples were decreased by around 25% and 50% respectively.

3.3 Toughening mechanism To investigate the toughening mechanisms, the fracture surfaces of the broken specimens after the impact tests were examined using a high resolution SEM. For the stronger nanocomposites (samples of agglomerate sizes <45μm and halloysite ball-milled for 2 and 14 hours), a lot of black holes are clearly seen at the edges of the crack surface, denoted as zones A and B in Figure 3.5(b). These black holes are micro-cracks which break-opened and reveal the wrapped agglomerates, as shown in Figure 3.7(a). In the mean time, there are

12

no such micro-cracks in the samples of neat PS, as Figure 3.5(a) reveals. A closer look at these micro-cracks, we can see that PS has flown into the voids of the agglomerates, and formed a matrix with them. When a micro-crack breaks open, the agglomerate debonds from PS, leaving PS ligaments around the agglomerate, as suggested by Figure 3.7(b) and (c). It is believed that the formation of these micro-cracks have absorbed extra energy, thereby increasing the impact strength of PS. During fracture, a crack propagates via breaking the agglomerates. Agglomerates can impede crack propagation, known as crack pinning. After the crack was temporarily pinned, the local stress would increase. When the stress exceeded the strength to fracture the agglomerates and/or the interface strength between the agglomerates and the outside PS region, fracture of the agglomerate and/or debonding of the agglomerate/matrix interface occurred [1]. B Direction of crack propagation Direction of crack propagation (a) (b) Figure 3.5 SEM photos of crack surface of: (a) neat PS, and (b) <45μm sample.

13

crack tip Figure 3.6 Further illustration of Figure 3.5 (a)

14

Halloysite. agglomerates (b) (c) (d) (e) Figure 3.7 SEM photos of crack surfaces of: (a) <45μm sample; (b) enlarged image of the micro-crack shown at the centre of (a); (c) enlarged image of another micro-crack shown in (a); (d) 2-hour sample, and (e) 14-hour sample.

15

If the agglomerates are too large, it will be a detrimental to the impact strength. On the other hand, if the agglomerates are too small, the agglomerates cannot stop the crack propagation.

4. Discussion 4. 1 Relationship between agglomerate size and impact strength In order to compare the improvement in impact strength in terms of agglomerate size on a common ground, mean diameters for each of the agglomerate sample were determined. For the sieved halloysite, the mean of the diameter range was adopted, which gives the following assumption:

Table 4.1 Converted mean diameter of different sieved halloysite particle size ranges Diameter range of 63-90 45-63 <45 agglomerates (μm) Converted mean diameter (μm) 76.5 54.0 22.5

For ball-milled samples, the diameters of the agglomerates were determined from SEM photos, Figure 3.6(d) and 3.6(e). It was found that the mean diameter of agglomerates on the crack surfaces of the 2-hour and 14-hour sample were 10μm and 3μm, respectively. Based on the arguments in 3.2, the following assumptions of mean diameters of the ball-milled samples were drawn:

16

Table 4.2 Converted mean diameter of ball-milled halloysite particles Duration of ball-milling (hour)

Converted mean diameter (μm)

Therefore, combining Table 3.3, 3.4, 4.1 and 4.2, the impact strength against agglomerate mean diameter can be presented, as shown in Figure 4.1 .

Figure 4.1 Impact strength against mean diameter of halloysite agglomerate (Error bars corresponding to 1 standard deviation).

A very impressive trend of impact strength against agglomerate size was discovered from Figure 4.1. The impact strengths were lowest at the two extremes of agglomerate mean diameters, while it reached a maximum at around 10μm, which is 3 times that of neat PS. It suggests that there is an optimal size of agglomerate which gives the highest toughening power.

17

4.2 Relationship between number of agglomerates and impact strength In fact, the impact strength also depends on the number of agglomerates. However, it could not be the dominant factor. In all of our samples, the weight percentage of halloysite was 5%, in other words, it can be assumed that if all the samples weighed 100g, then the weight of halloysite would be 5g. Let the density of halloysite be p, then according to the following calculations: Weight of halloysite Density of halloysite = Total volume of the halloysite particles 5 P = Total number of particles x volume of particle with mean diameter Consider the halloysite particles in the 2-hour sample, Total number of particles = 5 4pπ (mean diameter)3 3 2 From Table 4.2, mean diameter of 2-hour sample = 10μm 1.19 X 1015 Total number of particles = p Therefore, relative number of particles of the halloysite samples would be summarized by the following table:

Table 4.3 Relative number of particles of halloysite samples Sample <45μm 2-hour 14-hour Mean diameter of agglomerate (μm) 22.5 10 3 Number of particle x p 1.05 11.9 442 (X1014)

18

The impact strength of 2-hour sample is almost two times that of 14-hour sample, while the number of particles of the former sample is much less than that of the latter one. This reflects that the impact strength should not solely depend on the number of particles. However, if the number of particles is too small, the probability of stopping the crack propagation will be too low, and this leads to insignificant toughening effect. Therefore, both number and size of agglomerates will be affecting the impact strength.

4.3Patent filing Considering the rarity and successfulness of the procedure of synthesizing greatly toughened PS by blending with halloysite ball-milled for 2 hours, a patent search is currently in progress. The documents required for patent filing are also under preparation.

5 Conclusion Polystyrene nanocomposites with high impact strength were successfully prepared using halloysite as the only impact modifier. The process of the synthesis is simple, and inexpensive. In addition, halloysite, which is cheap, abundant and environmentally friendly [1], is the only filler and no other modifiers are required. With such readily available equipments and raw materials, the impact strength was effectively threefold of that of neat PS. SEM photos on the crack surface revealed the broke-opening of micro-cracks which formed a matrix with halloysite agglomerates. The toughening mechanism was

19

identified as the massive micro-cracking. Another highlight of this work is the discovery of the relationship between halloysite agglomerate size and impact strength. An optimal range of size was found which has demonstrated best improvement in impact strength. It is strongly believed that a further narrowing of this range can strengthen the polymer even more.

The significance of this work is recognizable. Firstly, it developed a new procedure which can toughen polystyrene very easily. Simultaneously, the filing of patent of this procedure is under consideration, and patent search is currently in progress. Secondly, the discovery of an optimal size range of halloysite agglomerate to strengthen polystyrene has contributed to the science of filler.

6 References [1] Yueping Ye, 'Microstructure and Properties of Epoxy/Halloysite Nanocompostie,' M.S. thesis, University of Science and Technology, Hong Kong, 2006. [2] Shiqiang Deng, Jianing Zhang, Lin Yea, and Jingshen Wu, 'Toughening epoxies with halloysite nanotubes', 2008 Elsevier Ltd. [3] Bourry D.; Favis B. D., 'Morphology development in a polyethene/ polystyrene binary blend during Twin-screw extrusion' Elsevier, Oxford, ROYAUME-UNI (1960) (Revue) 20