When I first heard from David Szego on twitter, I was intrigued. Keep tweeting at Elon Musk, David – he may get back to you some day but, in the meantime, everyone else can see it and start to wonder just what you’re up to! Read on for an interview with Mr. David Szego discussing the exciting idea of the Szego Mill.
Peter Bell: What can you tell me about how the Szego Mill can be used in gold mining?
David Szego: What we have is an industrial grinding mill. Anybody can buy a mill, and it’s an intrinsic part of the recovery process, but The Szego Mill has a few specific characteristics that no other kind of milling or grinding technology can offer. These will come into play when you are processing the ore to make the final product, which is quite a-ways down the line from the early exploration stages where you are, Peter, but it’s good to plan with ROI and efficiencies in mind as it can really change things in later stages.
We can talk about how to use the Szego Mill with graphite, and other minerals too. A key consideration with graphite is that you generally want it in a flake shape, but most grinding mills typically destroy that flake and make a round granular particle. Now, you do want graphite to be round for some purposes, but you often need to maintain that flake shape, which our Mill can do. Processing material in a way that doesn't destroy what you're trying to achieve in the end product is important because it means the processing becomes cheaper and easier – less steps to get what you're trying to achieve. I won’t get into the engineering detail with you here – we have 40 years of research papers for that, but that’s it really quickly at a high level.
Flake size isn't something you care about generally with gold, but there is something to discuss: There is a lot of black sand when dealing with placer gold and tailings. This black sand is called magnetite and it’s very hard, but there can be a lot of gold in that magnetite. It’s great to find a vein of gold, but there is still a lot of value in placer deposits and tailings to consider. Problem is, it’s somewhat difficult to recover. If you can approach recovery the right way, then these sources can be just as valuable as the vein.
When you grind up all that magnetite, you put it through a couple of chemical processes to get the gold out. We're able to do a couple of things if you grind the black sand in the Szego Mill. One is to take your chemical agents that you would normally use as a second step to dissolve the gold and to separate it from the magnetite, and add it right into the Mill as you feed the sand. If you do it all at once, then you're reducing the number of steps – less cost and time. As the sand is ground in the Szego Mill, it's immediately making contact with that dissolving agent before anything has a chance to oxidize or change state. The mass transfer rate is very high, and we find that you get 6-10% more leaching out of magnetite. That's 6-10% more gold recovery, as well as saving a step.
The other thing to consider is that when the gold particles separate from the magnetite in the Szego Mill, they get flattened due to the nature of our grinding action. The magnetite gets crushed into micron-sized granules, but gold is very soft, so it gets flattened and squished into a flake shape as it comes out of the Mill. There’s a huge advantage to this - when you try to separate the magnetite from the gold with typical sifting equipment, the magnetite falls through the mesh because it is smaller than the holes, but the now-larger and flatter gold flakes get trapped in the filters. You're not only extracting more gold quicker and easier, you're catching it as you do your particle separation from the magnetite. So now you have two advantages when recovering gold in the Szego Mill – two ways to get more of what you're trying to extract. I think it's worth looking at it when you start to design your processes.
Peter Bell: Always, David. It's a lot easier to just do the conventional thing as a Junior mining company, but a 6-10% boost is worth a closer look. And if there are opportunities for cost savings, too, then things are all the more interesting.
David Szego: I think it takes a guy like you on the team, Peter, who has that accounting background to assist the engineers and show the CFO. If they have someone who can say, 'Here are the numbers, take a look', then it can really help the overall return on investment.
Peter Bell: The strategic planning issues around building a mine are just huge. Very tough multi-disciplinary questions, too. Please, let me ask about another group in Newfoundland who are already in production. Very hard rock setting with small gold veins. Is this mill capable of doing what their ball mill currently does – grinding the hard rock down to the micron level?
David Szego: Not straight from the large ore. We take an input of about one centimeter and break it down to 10-20 microns.
Peter Bell: Would you still have some kind of a ball mill to do the preliminary grind to get it down to the centimeter size?
David Szego: Yes, there’s always a need for initial crushing when dealing with ore. You'd use either a hammer mill or a ball mill.
Peter Bell: Is the Szego Mill a type of a roller mill, generally?
David Szego: Yes, in a sense, but vertical. It’s shaped like a mug with three rollers inside it. The rollers rub against the inside of the “mug” cylinder, making them turn as the whole assembly of rollers spins. The rollers have grooves like a screw and the material gets caught in them, then transported lower and lower through the mill to the bottom as it is ground more finely. As the material is pushed up against the face of the thread or gap between the threads, there is a repetitive crush-release action. It goes “trap, crush, release, trap, crush, release” and so on until it makes its way to the bottom. That action is also what allows the dissolving agent to get mixed in and cause the chemical reaction.
Peter Bell: That sounds like a design where you can deal with some pretty hard material, as well.
David Szego: Yes, we use hardened steel for the rollers and the body, but we could use tungsten carbide coatings to deal with harder material. Or alternately food-grade steel for those uses. You don't have to worry about wear on the body because you could just polish it the way you refinish the inside of a car cylinder. It would just be a case of buffing it smooth.
The rollers are affordable enough that you would eventually replace them, really just the contacting sleeve on the roller. In a very hard material environment, you will get some wear but it doesn't significantly affect the performance of the Mill. When you're dealing with gold, the savings on processes and the increase on recovery makes the maintenance cost look like peanuts.
Peter Bell: Great. Interesting to think about breaking the grinding process down into two parts since I have heard that the ball mills can be very expensive when you’re trying to get a very fine grind. There is an exponential increase in energy costs with ball mills.
David Szego: Yes, it's ridiculously large. And a ball mill cannot do a thick, pasty-wet grind like ours can. You can't do that chemical reaction in the mill at the same time. Let me give you an easy comparison to a ball mill. If you were to put 10-20 tonnes an hour through a ball mill, then it would need to be about seven feet wide, twelve feet long, and ten feet high. You need to fill it with 25,000 pounds of steel balls, which would wear out quickly and have to be replaced because they're not hardened steel. You'd also need a reinforced concrete floor and it would take about 200 kilowatts an hour to spin the thing.
For the same capacity, a Szego Mill would be about your arm-span in diameter and you could drop it in place with a forklift. It would take about 30 kilowatts an hour and you could run it on a standard 30 horsepower motor, plugged into 550 volts. That size of mill would cost less than $250,000, whereas a ball mill would be double or triple that, plus installation, plus the steel balls, plus the power. It’s a massive difference.
Peter Bell: And the electricity for some of these projects isn’t just the marginal cost of the electricity – there may be costs to building high-voltage lines to site.
David Szego: Yes, think about installation on a mining site. And if you’re using gas for a generator, then things really start getting expensive. Power savings is extremely important.
So much has changed in the last ten years with chemistry, you can do away with cyanide and mercury now, but you still need the equipment to be able to do the chemistry. You can have a long process chain, but that is very expensive. The Szego Mill is a better option. It can easily remove a few steps, run cheaper, and cost less to purchase, install and maintain.
Peter Bell: How about testing capabilities? If somebody was to ship you some rock samples that had been ground down to one centimeter, then could you do some test work?
David Szego: Yes, in part. We’re setup to grind the material but don't have the equipment to analyze the output of the grind – it would have to go to a lab for that. Our clients often run it through their labs.
Peter Bell: Are you based in Ontario?
David Szego: Yes, We’re in an industrial space out near the Toronto airport, and a separate small office space.
Peter Bell: The family history is interesting. There are a few generations behind this.
David Szego: I’m glad you saw that! I managed to trace a patent for a ball mill back to 1898 in my Hungarian ancestry. It's crazy, but it makes sense… My great-great-grandfather invents this ball mill, then his nephew opened a metallurgy practice and farm equipment factory in the 1930’s, the sons picked it up and the brothers made their own inventions, then the next generation…
Peter Bell: Any idea when the shift from balls to rollers happened?
David Szego: 1960’s. The invention of the first Szego Mill started with a compressed spring. It’s there in the patents – if you take a hardened spring and you squish it down, you get a thread much like a screw. That was the original design and it probably would have self-destructed after about ten hours of use.
My great-uncle László Szegő invented it and was introduced to a professor at the University of Toronto, Dr. Olev Trass, late in life. They worked together for about a year and then my uncle passed away. The professor just kept going! For the last 40-odd years, he's just been working on it, improving it, publishing papers, doing commercial pilots, getting research grants, and more. When I became acquainted with him, we quickly became partners and he's passing the torch back to the family, as it were.
Peter Bell: Thank you, David. Inspiring to hear about the story of an idea that traces back over generations. It seems obvious that a mill using screws and gravity would work better than balls in a washing machine, but I’m not an engineer. There seems to be some excitement in the mining space these days around new ideas, so I hope the that time for the Szego Mill has come. I will be watching and try to help introduce the idea to various groups where I can.
David Szego: Thank you, Peter. Great speaking with you.
See more here http://www.szego.me/about/history-of-the-szego-mill/
