Printing with Wood

From RepRap
Jump to: navigation, search
                          Making Wood Polymer Composite (WPC) for Printing with Wood on a Reprap


Introduction

I want to print wooden objects. I make kinetic sculptures in wood and am really excited at the potential of additive printing. It seems so much more elegant to add bits of a material to make an object than to start with a block of the material and hack away at it until the object emerges leaving a pile of waste. However, there seems little available by way of suitable filament. I want to make my own, ultimately from my own recycled plastics and using a variety of different woods. Well it can't be that difficult, can it?

After some research I offer the following as a theoretical method. It’s been done many times in labs around the world but I’ve not tried it yet under shed conditions. I lay no claim to it, I’m just passing on what I’ve read and am going to give it a try.


Polymer choice

There are, obviously, two possible sources for the polymer, oil based or plant based. To be able to recycle plastic would be environmentally preferable as it re-uses plastic that would otherwise go to landfill and it doesn’t consume the extra energy needed to produce the bio-polymers. However recycled oil based plastics are not biodegradable or compostable, biopolymers are. It’s a choice we’ll all have to make for ourselves. The process is similar which ever polymer is chosen.


The Process

Thoroughly mix ground wood particles (wood flour/dust) and a powdered form of your chosen polymer and extrude into a filament. You now have a printable wood but it’s not that strong, you wouldn’t want to build a house with it. This is because wood and polymers repel like oil and water. The tiny wood particles are coated in the polymer but they are not bonding well. This makes the product a bit brittle and water absorbent so it breaks down over time as the wood particles swell and the polymeric structure flexes relative to the wood particles. This is good environmentally as it means your products can be composted or biodegraded. It’s not so good if you need a strong product. To overcome this you need to add what’s known as a coupling agent (also known as a compatibiliser or an interfacial modifier.) But these seem very difficult to come by in UK unless you want to ship 25kg from China. Mind you, that amount would make nearly 4 tonnes of wood composite if anyone fancies doing it on an industrial scale.


How to make WPC by not using recycled polymers.

1. Wood powder. This can be any wood. It can also be any organic fibres, e.g. straw, newspaper, card bamboo, coconut husk etc. It acts as a filler and the fibres add strength. It must be dry <3% moisture and have a particle size of less than 0.1mm (100 microns / 0.004 in / 140 mesh.) This is not as daunting as it sounds. Sawdust is typically 30 to 600 microns so get a 100 micron sieve. The stuff that comes off my belt sander is plenty fine enough. Obviously you need to match your particle size to your extruder nozzle size. If you are using a 1mm nozzle, you can go for a coarser powder. If you are using a finer nozzle go for a finer powder. Equally if you want a fine finish, go for a finer powder.

2. Polymer. In theory you can use almost any polymer that is within your extruder’s temp range but the most popular oil based polymer with commercial producers is Polyethylene (PE), which is a happy coincidence since my milk cartons are made from HDPE. You can also use LDPE, PVC, PP, ABS and PS. If you want to go the bio-polymer route then you can use the familiar PLA, or Poly-3hydrobutyrate (PHB) or Polybutylene Succinate (PBS). This can be bought in a powder form, ideal for our purposes.

3. Mixing. In short, they don’t like each other. Wood and other fibres are hydrophilic (they love water) whereas polymers are, like small boys, hydrophobic. It’s like trying to mix oil and water, or make a good mayonnaise. So you need a little help in the form of a coupling agent (CA.) You can make a wood polymer composite without a coupling agent. It’s a whole lot easier as getting hold of a coupling agent is nigh on impossible in UK. The disadvantages of not using a CA are that in the finished product the fibres can show some pull out making a smooth finish difficult and there is some loss of adhesion between the polymer matrix and the fibre particles giving a weaker product. Adding CA gives a 20% to 30% improvement in tensile strength and hardness. But given the difficulties and reading a number of academic studies on this issue my view is it’s not worth the extra hassle for the sort of products I want to print. Added to this, companies like Bioresins.eu market a variety of biopolymers ready mixed with all sorts of fibres.

4. Coupling agent (CA). If you decide on using a CA, whether or not you are using a conventional polymer or a newer bio-polymer, then usually Maleic Acid Anhydride is grafted on to the polymer which is then powdered and added to the mix later (see below.) However a superior agent in terms of strength is polymeric diphenylmethane diisocyanate (pMDI) but again I can’t find anyone willing to sell me any. I’d be happy to know if anyone has any luck on that.

5. How to mix. Simple. Just put the different dry powders into a plastic bag and give it a good old shake for 5 minutes.

6. Proportions without CA. Mix your polymer powder and wood powder at a ratio of up to 50/50 by weight. In many of the academic papers a 70/30 mix of polymer to fibre is recommended but this can be varied either way and experiment will reveal the best blend for your purposes.

7. Proportions with CA. If you want a heavier duty product, then use 2% by total weight of CA. Add this to a 70/30 mix of polymer/wood fibre by weight. So, for example, you have 70g of PHB and 30g of wood fibre so you add 2g of your coupling agent.

8. Extruding into a filament, (no CA.) After a good shaking, tip the powders into your filament maker. I’ve not done this yet in the shed, but in the papers I’ve read the suggestion is to run the filament maker at you polymer melt point at the hopper end and 15°C cooler at the nozzle end. The suggestion is that the length/diameter ratio of the extruder is 40 so a 15mm extruder auger needs to be 640mm long. I’m not sure if this is possible or necessary with a Lyman type design. Furthermore the materials need to be in the barrel for about 1.5 minutes so the auger speed will need to be adjusted. This will need experimentation to get it exactly right. The extrudate can be air cooled and wound onto a spool.

9. Extruding into a filament (with CA) It’s similar to extruding filament without a CA but reduce the temperature at the hopper by 10°C and at nozzle by 40°C. Keep materials in the barrel for 2 minutes minimum. Cool extrudate in air or a water bath then wind it onto a spool.

10. Printing. Feed the filament into your Reprap and print as usual but adjust the head temperature as needed to achieve a melt.



How to make WPC using recycled polymers.

Again, I’ve not done this yet but my research suggests it’s possible in-the-shed. You would need to shred your polymer as finely as you could and add your wood powder, give it a good shake then put into your extruder hopper as outlined above. If using a CA you need to add that too at the mix stage. The rest of the process is as suggested above.

How to make a grafted Coupling Agent

Mix 1 part (in powder form) of your chosen polymer with 2 parts (in powder form) of Maleic Anhydride (MA) , and 0.2 parts (in liquid form) of the initiator, 2,5-Dimethyl-2, 5-di-(tert-butylperoxy) hexane (known as Trigonox 101 in UK and Luperox 101 in US.) Vigorously shake for 5 minutes then put it in your extruder set up to melt your chosen polymer. Air cool the extrudate and shred it. If you can grind it to a powder then do so. Keep it in a airtight container.


Sources

1. http://gradworks.umi.com/34/95/3495950.html

2. Mechanical performance of biocomposites based on PLA and PHBV reinforced with natural fibres – a comparative study to PP

           A.K. Bledzki, A. Jaszkiewicz

3. Biodegradable polymers/bamboo fiber biocomposite with bio-based coupling Agent.

           Seung-Hwan Lee, Siqun Wang
           Tennessee Forest Product Center, 2506 Jacob Drive, University of Tennessee, Knoxville, TN 37996-4563, USA



Meerdog, October 2012