Features Research & Innovation
Thiosulphate and Paper Machine Corrosion: Is the Problem Solved?

The layman’s view of corrosion often amounts to “blame it on chlorides.” Winter road salt corrodes our cars, ocean freighters look rusty and everything seems to point to chlorides. On the other hand, fresh water has always been regarded as pretty innocuous — we use it to keep the car clean, wash off the salt, avoid the rust. The same kind of view used to prevail about paper machines — those on the coast might see more corrosion because of seawater-floated logs, but most machines were located on a river that supplied fresh water for papermaking without much corrosion. So it was something of a shock back in the 1970s when the industry realized that newsprint paper machines had become corrosive. Bronze forming wires could no longer be used, the drilled holes in bronze suction rolls seemed to get bigger by the month, 304L stainless piping sprung leaks overnight and new stainless suction rolls would suddenly crack and fail. All this mayhem happening in seemingly benign white water, where the only chemical addition might be a pinch of alum. Whatever got into that nice clean water to make it so aggressive? As you might have guessed from the title of this article, the culprit turned out to be a not-so-innocuous little anion called thiosulphate.

How do we know that thiosulphate caused this corrosion?

For a long time it was not really understood exactly why components of machines were failing. Was it wear or abrasion-related? Was it the chlorides? Were sulphate-reducing bacteria causing corrosion? Well, each of these things did happen sometimes, for some components on some machines. But the single most important problem was none of the above.

Corrosion testing is challenging: field-testing with coupon test racks is easy to put in place, but it takes time to get results, and interpretation can be deceptively difficult where conditions continually fluctuate. Accelerated testing in the laboratory attempts to go one step better using electrochemical instrumentation, but that introduces another form of complexity. In the end it took some carefully thought out, very straightforward laboratory jar tests with corrosion coupons to show that most of the material failures on paper machines were caused by thiosulphate.

A young engineer working at a mill in Quebec was the first to do the critical tests. In 1978, Marcel Lapointe, working for Rolland Paper in St-Jrome, demonstrated that small amounts of thiosulphate in distilled water could produce big pits in 304L after only two weeks (1)(Fig.1). Remember, in those days 304L was the standard material of construction for most machines, so this was an unpleasant surprise, just as unpleasant as the pits formed in white water piping (Fig 2). Later, the same kind of jar testing at Paprican showed that of the salts typically present in white water, sodium thiosulphate was the only one that could readily corrode bronze (2)(Fig.3). More to the point, it could knock out a couch roll (Fig 4).

These effects had been previously overlooked, partly because thiosulphate concentrations were not easy to measure in white water at the 10 ppm level. The polarographic technique was notoriously difficult to calibrate. Exactly why and how such dramatic corrosion could occur was not known at the time: no other industry had reported these effects. But there was no doubt that thiosulphate was the culprit.

Conventional wisdom of the 1950s and 1960s held that bronze wires were in fact degraded by a wear mechanism. Certainly wires did become worn when the bronze surface of a couch roll was roughened by corrosion (Fig 5). Rubber covers on couch rolls helped somewhat, but wires still failed all too quickly. In those days many machines were shut down each weekend for wash-up, so a full-blown crisis was precipitated when the wire would not remain intact between these weekly shuts. The bronze wire-life crisis of the fifties and sixties led to the development of the non-metallic forming fabrics still used today. These fabrics made from polyesters and nylons have excellent wear resistance and, guess what, they are unaffected by thiosulphate.

As for those cracking suction rolls, so many rolls failed in the 1970s that it took quite some time to unravel the complexity of the problem. New stainless rolls had been introduced as improvements over bronze but they were unexpectedly susceptible to cracking (Fig 6). Was it the casting or drilling technique? Was it the alloy formulation or heat treatment? Was it a roll loading or design problem? What about residual stress in the roll? How come the new stainless alloys worked fine for ship propellers in seawater but cracked in seemingly innocuous white water? We now know that chlorides were not really needed for corrosion and cracking to occur: thiosulphate has some nasty ways of breaking down protective oxides all by itself.

Where did the thiosulphate come from?

Thiosulphate came from hydrosulphite brightening. The Powell River Company was one of the earliest users of hydrosulphite as a brightening agent in the 1940s. Zinc hydrosulphite helped them make newsprint from the darker coastal wood species. When inland mills started to use brightening, adverse environmental effects of zinc in the effluent were identified and the industry switched to sodium hydrosulphite. This salt is significantly less stable than the zinc salt in the mildly acidic water then commonly used for newsprint manufacture: it has a tendency to break down, undergoing a time-dependent disproportionation or “self-reduction” reaction at pH 4.5. One of the decomposition products is thiosulphate. So a change in brightening agent had introduced a new anion into the machine white water, one whose corrosiveness was unexpected and remained unrecognized for a long time.

Exactly where is thiosulphate produced?

Sodium hydrosulphite can be supplied to the mill as a powder or liquid solution with stabilizers and chelating agents, or it can be made on-site from sodium borohydride. Occasionally the as-generated brightener can contain significant levels of thiosulphate. More commonly, thiosulphate is created during storage, particularly when hydrosulphite is stored unrefrigerated, for longer times, or with insufficient alkali present. For example, in the summertime warm weather and warm make-up water can promote thiosulphate formation during hydrosulphite storage (3).

Pulp brightening is usually done in a tower or chest where lower pH and excessive application rates (e.g. over 10kg/ADT) can create thiosulphate (4). Higher application rates and trim brightening just before the head box can stream hydrosulphite onto the machine, where it may decompose and form thiosulphate.

Are there any other sources of thiosulphate?

Newsprint used to be made with low-yield sulphite reinforcing pulp. Over-cooking the pulp could produce thiosulphate, which then finds its way onto the machine. In theory, the more modern chip-sulphonation processes might produce thiosulphate, but this has not been commonly observed.

What is different today?

* Better hydrosulphite practices have been adopted.

In the 1980s the industry was quick to implement much better practices for hydrosulphite generation, storage and brightening as outlined above, so as to minimize thiosulphate contamination of white water. Better white water sampling and analysis techniques were also developed so that good practice could be verified. A recommended maximum concentration of 5 ppm thiosulphate was identified for 304L machines as a working guideline. Those machines that were unable to consistently meet this level gradually upgraded their way out of the problem with 316L piping and better stainless suction rolls. The fact that some 30-year-old bronze couch rolls are still running on machines that use hydrosulphite is evidence enough that, with diligent attention to detail, good hydrosulphite practices can be consistently successful in controlling thiosulphate corrosion.

* Better construction materials are now in use.

Over the past 25 years, 316L has become the standard material of construction for almost all paper machines, including newsprint machines. 316L resists thiosulphate pitting for white water temperatures of at least up to 60C. We are now seeing the introduction of a new family of duplex stainless steels as the next generation which may eventually take over from austenitic stainless steels like 304L and 316L. New steel making technology now allows the routine use of nitrogen as an alloy component in 2304 and 2205, making these duplex steels more pitting resistant and stronger. They have the added advantage of resisting stress corrosion cracking that can occur on higher temperature components such a steam boxes. Gary Mills of Encore Metals in Delta BC puts it this way: “New duplex stainless steels like 2304 and 2205 are now being used much more extensively in pulping and papermaking. They can readily handle the chemistry employed in today’s newsprint white water systems.” Paul Glogowski of Metso Paper has written a useful summary of the materials currently used to build paper machines (5).

Premature cracking of suction rolls is now pretty much a thing of the past, mostly because better alloys are now available for these components too. One stainless steel developed in 1986 has proven to be a real workhorse. Greg Michel of Sandusky International commented: “The current generation of cast duplex stainless steel suction rolls has proved very reliable and corrosion resistant. Sandusky Alloy 86 is commonly specified for both couch and press positions in modern high-speed paper machines.” Another company has had good success with duplex stainless rolls that are weld-fabricated, and then very carefully heat-treated in a vertical position to minimize residual stress in the roll. Consultant Ralph Davison noted: “Outokumpu Prefab’s 3RE60 welded stainless steel suction press rolls, with over 1300 rolls in service, have been very successful in machines across Europe, North America and elsewhere, even for the most critical applications.” The company has also supplied a number of suction rolls in welded 2304 plate for the less demanding couch position. Max Moskal, a consultant with Mechanical and Materials Engineering, has written a useful discussion of the considerations surrounding suction roll selection which includes lists of alloys past and present, and also references a number of roll inspection guideline documents (6).

Has thiosulphate corrosion gone away for good?

Many more wet-end additives are used in newsprint manufacture today as compared to 25 years ago. Additives are used for control of formation, retention and contaminants and for the most part they do not influence white water corrosivity. Some other chemistry changes are indeed beneficial. Neutral papermaking is a term that refers to the use of precipitated calcium carbonate (PCC) in the mechanical pulp furnish. The switch from the traditional pH 4.5 to pH 7 in itself makes little difference from a corrosion viewpoint. But the PCC introduces a significant level of bicarbonate into white water and that will tend to inhibit corrosion of 304L, though not bronze. Some mills deliberately add sodium bisulphite for its effect on pulp brightness. This too will inhibit pitting in 304L.

New sources of hydrosulphite, safety-related pressure to avoid in-mill use of liquid SO2 and a general push to reduce chemical costs are all recent changes that could impact thiosulphate levels. Many older machines still run bronze couch rolls and have some 304L piping or chests. Continued attention to good hydrosulphite practice is essential, both to avoid thiosulphate corrosion and to ensure that all the hydrosulphite fulfills its intended purpose of brightening mechanical pulp.

Conclusion

A combination of better hydrosulphite practices and better materials of construction was successful in controlling thiosulphate corrosion in newsprint paper machines in the 1980s. Today’s processes can still contaminate white water with thiosulphate. The challenge is to ensure that hydrosulphite brightening and associated procedures remain optimal, particularly with older machines where materials of construction have not been fully upgraded.

References

(1) M. Lapointe, O.Stitt, L.H.Laliberte, L.Belanger, R.D.Cloutier, Pulp and Paper Canada, Vol. 80, No.8, p.79, 1979.

(2) A.Garner, Pulp and Paper Canada, Vol.86, No. 3, p.T62, 1985.

(3) A.Garner, Pulp and Paper Canada, Vol.83, No. 10, p.20, 1982.

(4) A. Garner, J. Pulp Paper Science, Vol.10, No.5, p.J51, 1984.

(5) P. Glogowski, p.98 in “Stainless Steel and Specialty Alloys for Modern Pulp and Paper Mills”, Ed. A.H.Tuthill, The Nickel Institute 2002, www.nickelinstitute.org

(6) M.D. Moskal, p.112, reference (5).

Andrew Garner is principal of Andrew Garner and Associates Inc. He has 30 years experience in corrosion and materials engineering. He can be reached at andygarner@telus.net