Is a picture worth a thousand PhDs? Reflections on how we solved the puzzle of bacterial leaching

A picture is worth a thousand words!

Apparently.

Or is it?

Many times, pictures are used as evidence. Photos of the speeding vehicle, CCTV of the criminal act, micrographs of the metal fatigue, the list goes on and on...

Look at this picture…

I am going to tell you how this little picture was the first clue in new scientific understanding of nature. ?It was shown to me as evidence that bacteria directly attacked minerals. The little holes are about a micron in size, the same size as the bacteria. It was obvious that bacteria have little bacteria ‘teeth’ that were chowing down into the mineral from its surface.

Now before you stop reading because you think that this is a niche area of science undeserving of your attention, take a moment to reflect that this is going on in your mouth as we speak – not the same bacteria, and not the same mineral, but a major factor in tooth decay is the action of bacteria on the mineral surface of teeth. And don’t forget the environmental issues that acid mine drainage causes!

?‘Yes…..but…..!’

When I first saw this picture and other kinetic evidence, I agreed but was mildly sceptical. It wasn’t my area of expertise. My claim in the world is that I have pioneered a theory of dissolution based on the combination of semiconducting nature of some minerals and the electrochemistry of corrosion. Physics, and physical chemistry. Far away from biology and life. So, I accepted, with the drawn out “Jaaaa, okay” of South Africans, translated means ‘yes, buuut…!’

At the time, I was at the University of the Witwatersrand , Johannesburg, and my research was funded by Gencor, a major mining company at the time. I modelled carbon-in-pulp tests results for them as part of their campaign to develop new gold mining opportunities. Part of Gencor’s research effort was focused on BIOX, their bacterial leaching technology. They led the world with the first commercial bioleaching plant at the Fairview gold mine in Barberton. At about this time, they added a bacterial reactor in front of an autoclave at the Sao Bento gold mine in Brazil to increase the capacity of the plant, and then they built a huge BIOX facility in Ghana.

As part of their BIOX process development (now Metso Outotec ), they wanted a computer model to investigate opportunities to optimize their technology. Dave Dew from Gencor approached me: “Can you build a model based on a fundamental understanding of the BIOX process? As a company, we favour the direct mechanism. I suggest you start by giving us a literature survey of current models. When can you give us a report?”

Even though this was thirty years ago, the volume of research that had been conducted was phenomenal. Debbie Miller, also of Gencor, had done an MSc with Geoff Hansford at 南非开普敦大学 , and had a pile of research papers that made lawyers look quite unprepared for their day in court!

My task: read, comprehend, make maths!

The central issue - do bacteria have teeth?

I uncovered two things. First, the researchers were still fighting about whether or not bacteria had ‘teeth’, that is, is there a ‘direct’ biological mechanism for dissolving mineral sulphides. Or is the bacterial action ‘simply’ due to their ability to oxidize iron really rapidly and it was the oxidized form of iron (ferric ions) that did the work on the mineral? ??

This debate was a bit complicated to unravel. The bacteria attach to the mineral surface, which seems to support the ‘direct’ mechanism. Combined with pics like that in Figure 1, you would think it was a slam-dunk in favour of ‘direct’ bacterial action.

Except that all bacteria seem to prefer to attach to surfaces (think of plaque, which is formed by bacteria on your teeth). And isn’t there the possibility that bacteria even on the surface are ‘simply’ iron-oxidizers? So, this was the debate, illustrated in Figure 2, apparently unresolved for 35 years since the discovery in 1947 of these bacteria.

Second, someone, I forget who, had tried to grow these bacteria in an electrochemical cell. Wow, now this enters my world, the world of electrochemistry!

What the maths taught me…

I am not a mathematician, although I wish I was… but I have found that writing down the equations and solving them, mostly on a computer, creates clarity of thought that cannot be achieved by ‘thinking it through’. These systems have too many variables. Besides, the information in these systems is about rate, ‘how fast did this reaction go?’, which implies that there might be different paths, and the question is not whether a path exists or not, but whether is it the one that is used, the one that is dominant. So, I wrote down the equations and tried to solve them.

I was just thinking about these equations at the beginning of the summer holiday, about December 1993, when Prof David Glasser asked me to contribute to a special edition of the Chemical Engineering Journal (CEJ) . I spent the entire summer grappling with equations, computers, programming, data and mechanisms, and submitted my first bacterial leaching paper.

This initial model was a lot harder than I thought, and I spent the entire summer working at the start of the academic year, I bumped into my colleagues Charley Cooper and Steve De Kiewit in the Senate House coffee shop, a veritable melting point for ideas across the University over a insipid drink. I was gob smacked that he spent the time sailing up the Mozambique Channel!

Seeing my envy, he smiled, took a puff from his pipe (indeed!), and told me that as they were passing Richard’s Bay, he saw all these families on the beach, play and enjoying themselves in the sunshine and waves, and he was envious. To be with family, and really relaxing, while he had the task of delivering this yacht to the Seychelles with the possible threat of summer cyclones down the Channel. And before he could finish his musings, a powerful cigar speedboat pulled up next to the yacht, and after the pilot had found out his mission to the Seychelles, he begged Steve to take him along! And think about it, the parents on the beach were probably wishing they had the cigar boat! A triangle of jealousy!

Believing is seeing

Steve also introduced me to rock art of the San people. David Lewis-Williams had made a bold hypothesis that much of the art is of a religious nature and should be interpreted in terms of trance states that were entered into due to extended dancing during religious ceremonies. Lewis-Williams’ thesis is titled: “Believing and seeing”. We usually claim that seeing is believing, arguing that the hard facts prove the point.

But the opposite is true. Steve and Lewis-Williams taught me that our beliefs preselect the facts. “Believing is seeing”. So, if we believe that these bacteria have teeth, we will preselect the picture with holes and claim this as the evidence we need. We will add the knowledge the bacteria love to attach to surfaces, and QED, we have the preeminent facts. We do this all the time, believing is seeing!

Where’s the murder weapon, Piorot?

To prove the ‘direct’ mechanism you need to be able to identify the ‘teeth’. What biological component are the bacteria making that enhances the rate of dissolution? People had tried to identify such a component: is it hyaluronic acid? Or a component in the polysaccharide layer around the cell or in the biofilm? Or something else… People had tried to find something, anything, to pin their beliefs to.

Experiments had been set up in which researchers added bacteria and minerals to their shake-flasks and tried to limit the amount of iron. If there is no iron, and still dissolution, then it is biological, they have ‘teeth’. These results supported the direct mechanism.?

Unfortunately, there is always some iron present, whether in the culture (because we culture the cells in an iron-rich environment) or from the mineral itself. My models of these shake flask-type experiments, eventually presented at a conference in Montana in 1996, showed that you need little iron to get the indirect-type pathway going. Low iron cases simply made the ‘lag’ phase longer. Of course, this isn’t a real lag phase - a real lag phase is one in which internal biological phenomena are being reactivated in the organisms. In this case, it is simply a lack of food, iron, and this shortage is slowly overcome as iron dissolves from the mineral sample.

No smoking gun! No murder weapon! Worse, without identifying the biological teeth, researchers could not demonstrate any causal connection. An unsolvable problem, in a sense.

Not a scientific question!

If you think about it, the assertion that “bacteria directly attack the sulphide minerals” is not a scientifical testable hypothesis with a clear causal link. One needs to identify the causal agent and rearrange the statement to something like “bacteria secrete prA when attached, and prA enhances the rate of oxidation”. This statement can be tested – for example - create genetic mutants without the ability to produce prA, and demonstrate bioleaching doesn’t occur, thereby establishing the causal link.

But the former statement, that “bacteria directly attack the sulphide minerals”, is not scientific – and this was where the community was stuck.

I realized that the scientific question cannot focus on an unidentified chemical. When I got to this point, I knew exactly how to restructure the question. Instead, I focused on the role of iron: “If there is a direct mechanism of bacterial attack, increasing the level of iron will not increase the rate of leaching.” This is testable.

‘No ideology’ tests

A lot of science is a “let’s see what happens if?” investigation. For example, what happens if I mixed this with that? The primary question is “what happened?”, and the secondary question is “can I explain it?” This type of investigation, ‘discovery science’, is important, and is the more rewarded area of science. But it doesn’t follow the scientific hypothesis-testing approach – on its own, it is information gathering not hypothesis testing.

Much of the data obtained during “what happens if?” science is open to interpretation and is plagued by “believing is seeing”. We only see and report the results that fit our beliefs.

Now that I had reformulated the question, what was needed was tests that could test the hypothesis without ideology.

“Without ideology, what’s that”? Ideology is our set of beliefs. For example, I once was talking to an executive for technology from Anglo American , and he told me, half-jokingly, that he would never install a bacterial leaching plant at one of their mines because “I don’t believe in taking advantage of the sex lives of organisms!” Another ideology: pressure leaching is easier to run than bacterial leaching. Notions that are baked-in as preselected viewpoints with a rigid consistency, these are ideological views.

Obviously, the ideologies of scientists are subtle. My experiments needed to be able equally show that there was a possible direct mechanism ('teeth') or a possible indirect mechanism (simply iron-oxidation).

Electrochemistry to the rescue!

What I proposed to Dave Dew at Gencor (now BHP ) was this, that I do experiments with and without bacteria at different concentrations of iron. However, because iron is present in both oxidized (Fe3+) and reduced (Fe2+) forms and their concentrations change through the course of the test, it can be hard to detect a bacterial effect with certainty.

Crucially, the concentrations of the oxidized and reduced forms of iron had to be kept constant during the course of each test. In other words, keep the redox potential constant during the test.

Such constant redox potential tests were known in hydrometallurgy, and I had published a study where I had added diluted peroxide to the vessel to maintain the redox potential. However, peroxide and other chemical oxidants would kill the bacteria. How could I do this?

To the rescue did electrochemistry come, as a Shakespeare might have said! So, my proposal to Gencor was that we do the experiments in one side of a divided electrochemical cell and use the electrochemical current to keep the oxidized and reduced forms at a constant pre-determined value. This is shown in Figure 3.

A radical experiment

This experimental design was, and still is, thoroughly radical!!

My proposal was that we would do the tests at different concentrations of total iron and keep the ratio of Fe3+ to Fe2+ constant throughout each individual test. Comparing the results with and without bacteria at exactly the same values of total iron and ferric and ferrous forms of iron, would reveal the role of bacteria.

Dave Dew realized what I didn’t: this is a big programme and should be broken into achievable parts. The first part would be to develop the experimental apparatus, and that he felt, Gencor would fund.

What followed was five years of work with several PhDs and MSc students.

Midway through this programme, I was invited to the Netherlands as the external examiner for a PhD on bacterial leaching. Sitting at a pizzeria, the Dutch professor brusquely said: “You know, I think that this whole bacterial leaching area is a nice little question for mediocre scientists.” Booyakasha! as AliG would say.

The insult still stings, even though I could have pointed to logical error he and his student had made. They had used the exact same data to derive both the rate of chemical leaching and the rate of bacterial growth. When these two sets of results were the same, they then amazingly concluded that the mechanism was the same! I kept quiet, and the next day I passed the student…?

Making it happen

1994 was a difficult year. Paul Harvey started on the project in my lab along with Paul Holmes, Paul Knottenbelt (research group of Pauls!) and Les Bryson . At the same time, the University sent me to Implats (Impala Platinum) ‘to some industrial experience’. Paul Harvey did well in a semi-vacuum!

Midway through that year, I went to an amazing Engineering Foundation conference in Snowbird, Utah, focussed on bacterial leaching. There I met Bill Costerton of the Center for Biofilm Engineering (CBE), and then convinced GOLD FIELDS to fund me for a trip to Montana at the end of that year after I had finished at Impala.

The CBE had a wonderful social atmosphere, unlike what I experienced at other organizations, and the two months there were filled with fun and new people.? I studied the growth of biofilms of these organisms on pyrite using their confocal laser microscope.

I used pyrite that my dad got for me from the Grootvlei Gold Mine in Springs, where he worked underground. The pyrite was sliced and mounted in small flow through chamber with a cover-glass. Using the confocal features of the microscope, I observed the structure of a biofilm of these bacteria on a mineral for the first time. The pics I took are shown in Figure 4. What was interesting was that the bacteria were more concentrated towards the solution side of the biofilm!

What does this mean?

No bacterial attack

The next few years were a blur of activity: Terry Fowler, Darryl Howard and Yvonne Driessens took over from where Paul Harvey finished. ?Paul Holmes joined to look at the electrochemistry of dissolution, while other areas of my research group increased ( Jurgen Gnoinski , Douglas Mughogho , Zahed S. ).

By the end of 1998, we had conducted the most tightly controlled leaching experiments ever done. And the first results were in, as shown in Figure 5. Clearly, there was no direct bacterial effect. The amount dissolved with bacteria was the same as that dissolved without bacteria. If they are the same, the role of bacteria is not direct, but indirect, as an oxidizer of iron!

Clear, conclusive, unbiased.

Paul Holmes’s detailed electrochemical work dovetailed excellently with Terry’s leaching work. Together, we extended the conditions and looked for contra-arguments. All roads led to the same conclusion: there is no direct mechanism of bacterial attack.

Hardcore science is actual knowing

In some senses this result is less interesting to the world, and in a world where interesting outweighs rigorous, we were bound to get a ‘meh’ response. Or maybe it is a mediocre area, as the Dutch professor asserted. Maybe I should have been more ambitious. But, as Steve De Kiewit taught me, there is a triangle of envy…

At the age of ten, my younger son, Anthony, remarked on how amazing it must be to know something that no-one else in the world knows. This is the joy of science: knowing.

With bacterial leaching, my students and I knew, and knew first! Others claimed, and perhaps got more recognition, but their claims were speculative or flawed. We really, really knew, we grokked it, first... And that is what hardcore science is about, actual knowing.

As a hardcore scientist, look again at Figure 1. Look at the shapes of the holes that the bacteria have supposedly dug? Hexagonal holes? Is that really evidence in favour of direct bacterial attack? Or in favour of chemical ‘etching’? Believing is seeing!

About the author:

Frank Crundwell, founder of CM Solutions Metallurgical Consultancy and Laboratories has been awarded the Milton E Wadsworth Award of the Society for Mining, Metallurgy & Exploration Inc. (SME) , and is an international member of the prestigious US National Academy of Engineering . He is the author of Finance for Engineers and co-author of Extractive Metallurgy of Nickel, Cobalt and Platinum Group Metals and is a leading authority on leaching and dissolution of minerals and salts.

?

Look it up

1.?????? Harvey, P.I., Crundwell, F.K. Growth of Thiobacillus ferrooxidans: A novel experimental design for batch growth and bacterial leaching studies. (1997) Applied and Environmental Microbiology, 63 (7), pp. 2586-2592.

2.?????? Crundwell, F. The formation of biofilms of iron-oxidising bacteria on pyrite. (1996) Minerals Engineering, 9 (10), pp. 1081-1089.

3.?????? Harvey, P.I., Crundwell, F.K. The effect of As(III) on the growth of Thiobacillus Ferrooxidans in an electrolytic cell under controlled redox potentials. (1996) Minerals Engineering, 9 (10), pp. 1059-1068.

4.?????? Fowler, T.A., Crundwell, F.K. The role of Thiobacillus ferrooxidans in the bacterial leaching of zinc sulphide. (1999) Process Metallurgy, 9 (C), pp. 273-282.

5.?????? Fowler, T.A., Crundwell, F.K. Leaching of zinc sulfide by Thiobacillus ferrooxidans: Experiments with a controlled redox potential indicate no direct bacterial mechanism. (1998) Applied and Environmental Microbiology, 64 (10), pp. 3570-3575.

6.?????? Fowler, T.A., Holmes, P.R., Crundwell, F.K. Mechanism of pyrite dissolution in the presence of Thiobacillus ferrooxidans. (1999) Applied and Environmental Microbiology, 65 (7), pp. 2987-2993.

7.?????? Holmes, P.R., Fowler, T.A., Crundwell, F.K. Mechanism of bacterial action in the leaching of pyrite by Thiobacillus ferrooxidans an electrochemical study. (1999) Journal of the Electrochemical Society, 146 (8), pp. 2906-2912.

8.?????? Driessens, Y.P.M., Fowler, T.A., Crundwell, F.K. A comparison of the bacterial and chemical leaching of sphalerite at the same solution conditions. (1999) Process Metallurgy, 9 (C), pp. 201-208.

9.?????? Fowler, T.A., Crundwell, F.K. Leaching of zinc sulfide by Thiobacillus ferrooxidans: Bacterial oxidation of the sulfur product layer increases the rate of zinc sulfide dissolution at high concentrations of ferrous ions.(1999) Applied and Environmental Microbiology, 65 (12), pp. 5285-5292.

10.?? Howard, D., Crundwell, F.K. A kinetic study of the leaching of chalcopyrite with Sulfolobus metallicus. (1999) Process Metallurgy, 9 (C), pp. 209-217.

11.?? Crundwell, F.K. How do bacteria interact with minerals? (2003) Hydrometallurgy, 71 (1-2), pp. 75-81.

12.?? Crundwell, F.K. Modeling, simulation, and optimization of bacterial leaching reactors. (2000) Biotechnology and Bioengineering, 71 (4), pp. 255-265.

13.?? Fowler, T.A., Holmes, P.R., Crundwell, F.K. On the kinetics and mechanism of the dissolution of pyrite in the presence of Thiobacillus ferrooxidans. (2001) Hydrometallurgy, 59 (2-3), pp. 257-270.

?

?