A thorough understanding of the flotation process could help reduce the cost and complexity of recycling systems

by Roland McKinney

A better insight could help flotation technology take off

Flotation applications are found in industries as diverse as coal fines recovery, contaminated soil treatment, gold ore thickening, minerals processing and paper recycling. All of these uses developed as a result of the discovery, in 1903, that air bubbles would collect chemically coated mineral particles in a weak acid solution. This was called froth flotation and led to the development of the process for ore dressing. It was not until the early 1930s, however, that flotation was applied to deinking. A patent was granted in 1935 to Hines and one of the first reviews of flotation deinking appeared in 1933.

But it was not until 1952 that the first flotation deinking system was installed - at a paper mill in the USA. Europe had to wait for almost another 10 years for its first installation, at a tissue mill in Greece. The first cells installed in a paper mill were taken directly from minerals processing. These were low capacity Denver "Sub A" cells, installed at a pilot scale trial. The units were open, rectangular vats with mechanical removal of flotation froth by a rotating paddle and mechanical mixing of air and stock.

Since then, there have been many changes in cell design. Voith started a development program in 1956, with its first installation in 1960. In the ensuing years other suppliers entered the market, but for about 20 years there were few changes to the basic design, which was still similar to the Denver cell. The following years, however, saw significant changes in flotation cell design as many more suppliers entered the market.

Radically different designs continue to be introduced and the 1990s saw the introduction of the gas sparged cyclone by Ahlstrom Kamyr in 1991 - redesigned in 1995 as the contactor - the flotation column by Kvaerner in 1996, along with numerous evolutionary changes by other suppliers such as Beloit, Comer and Lamort.

Without flotation deinking, it is unlikely that deinked pulp quality standards would meet the demands imposed by today's high speed paper machines

But despite these design changes, improvements in performance are not always obvious. For example, in some cases Voith "paddle cells" (essentially designed in the late 1950s) are operating beside "new" cells on the same or similar stock, and are giving a similar performance in terms of ink removal and brightness increase, to the more modern design.

Although new entrants to the flotation deinking equipment market have increased the choice available, mergers and acquisitions have resulted in fewer vendors and modifications to some designs. Mergers included Voith with Sulzer Escher Wyss to produce Voith Sulzer and Black Clawson with Lamort/Fibreprep to form Thermo Black Clawson. As a result of the Voith Sulzer merger, the Voith elliptical cell was equipped with the step diffuser inlet, which was said to improve efficiency through the removal of a wider size range of ink particles.

Deinking deluge

Voith Sulzer was not alone in changing the flotation cell, however. There are now many technical variations, including varied size, shape, air injection method and depth, froth removal and so on (Table 1).

Figure 1 - Growth in Installed Flotation Deinking Capacity

Without flotation deinking, it is unlikely that recovered paper quality standards would meet the exacting demands imposed by today's high speed paper machines. Recycled fiber would be used in less demanding applications and at lower proportions than is now the case. Its importance is illustrated by the growth in installed capacity (Table 2 and Figure 1).

As surveys of installed capacity tend to build on earlier estimates which include systems that may have been shut down or replaced, later years are probably overestimates of actual operating capacity. But the growth trend is clear nonetheless. Each decade has seen very significant increases in capacity, expressed either as the average increase in tons/day or in percentage terms.

Flotation capacity in both the USA and Europe increased dramatically in the 1990s, so there are few geographic regions in which "very" rapid growth rates are now possible. Over the next decade capacity will continue to grow, but at a lower percentage rate than last decade.

A factor limiting future growth will
be the availability of recovered paper. Although this prediction has been made many times in the past, high levels of
used paper recovery have been reached in countries such as Austria, Germany,
Japan and the USA. Further increases in recovery are becoming more difficult to achieve and may depend on legislation. Nonetheless, it is likely that flotation deinking will have a faster growth rate than paper production for many more years to come.

In pre-flotation, the brightness increase is normally greater than 10% ISO, whereas in post-flotation increases are typically less than 2% ISO. The latter is primarily used as a cleaning step and no chemicals are added. This is to enable residual chemicals to be removed, such as soap. Another difference is in the pH. As most European newsprint machines operate at or near neutral pH, post-flotation is usually in the pH range 6.0-7.5, whereas pre-flotation is in the range 7.5-9.5. Any organic material that is agglomerated by the pH change from alkaline to neutral conditions can be removed by post-flotation.

As post-flotation follows dispersion, stickies are reduced in size and there is a widespread belief that small (micro) stickies are removed by the post flotation step. Since there is usually an accompanying thickening stage, those that are not removed by flotation can be removed by treating the filtrate from the thickening step. This is the basis of a two loop recycling system.

One of the most important areas in newsprint flotation deinking is the role of old magazines or ash. It is normal practice in Europe to deink a blend of old newsprint and magazines, usually with magazines contributing 30-50% of the furnish. As this is a standard furnish the magazines and newsprint are collected together and in general are only available separately as over-issue grades.

In the USA, old newsprint and magazines are supplied separately and blended at the mill. The use of a blend has become standard in Europe because there is widespread belief that the presence of magazines considerably improves the efficiency of flotation. Yet there is not a proper understanding of the role of magazines. A consensus of opinion distilled from several mills is that the presence of magazines stabilizes foam resulting in less need for collector chemical.

But in the face of this standard commercial practice there are some who question the belief that magazines improve flotation efficiency. Setting aside industry practice, data on this topic in technical literature is divided and no clear model for the role of magazines (if any) has emerged.

Confusing chemistry

One aspect of magazine use that has been investigated is the addition of fillers, as these are present in magazines. The potential role of other components from magazines has largely been ignored - for example chemicals leached from fibers or inks or binders used in coating formulations. Many of these have surfactant properties and would be expected to influence the surface properties of ink particles.

One problem is that laboratory results suggesting that magazines do not increase flotation efficiency may not reflect a typical mill situation. In a detailed review, it was suggested that differences between laboratory and mill results were due to the presence of "natural" surface active agents in mill water systems, which are not present in laboratory systems. Also considered to be important was the ability of laboratory equipment to remove these by adsorption on to clean surfaces, unlike the mill situation.

Even if this hypothesis is not correct, it is clear than the chemistry of mill flotation is much more complex than that of laboratory flotation.

In another laboratory study comparing the effects of ash from magazines or added fillers, no beneficial effects from magazine ash or filler addition were found. Brightness after flotation of newsprint with 5% or 10% ash supplied by magazines was no higher than from flotation of newsprint on its own. But it would be incorrect to extrapolate these results to a mill situation, especially as some of the reported results were very different from industry norms.

Brightness before flotation was reported to be in the range of 45-47% ISO, depending on the proportion of ash (magazines). These brightness levels are similar to those seen in a typical mill. Following flotation, however, the brightness values were in the range of 49-51% ISO, an increase of some 4% ISO. These brightness increases are much lower than normally seen in mills, where brightness gains reach between 10-14% ISO, giving brightnesses after flotation of typically 55-58% ISO. The point is that the laboratory tests did not reflect a typical mill situation, so their conclusions apply only to the laboratory system examined.

In another study, it was concluded that there was no evidence from their results that ash components from magazines facilitate the removal of ink from newsprint. In this case the brightness after pulping fell with increasing magazine content. However, the reported brightness after pulping of newsprint (with no magazines) was approximately 60% ISO. This level is very high and is generally the final target brightness for newsprint mills after flotation and bleaching. Another failing was the method used to prepare sheets for brightness measurements, which was the same as for hand sheet preparation. This type of method has been shown to be inappropriate when measuring deinked pulp brightness.

Yet again the laboratory simulation had no relevance to an industrial situation and its conclusions can not be confidently extrapolated beyond the laboratory system. The problems of relating laboratory studies to the industrial scale is difficult to solve and there remains much to be learned about the role of magazines/ash/extractives in newsprint deinking.

Stickies solution?

As flotation removes hydrophobic ink particles and stickies tend to be hydrophobic, this would suggest that stickies removal across flotation should be possible. In fact some vendors of flotation cells suggest in their literature that this is indeed the case. However, when stickies removal in several flotation systems, for both newsprint and office papers was assessed (by the author), no statistically significant evidence for the removal of stickies was found. This should not be too surprising. The removal of large ink particles during flotation is widely recognized as being inefficient, so the removal of other large contaminants is unlikely. However, in some systems the surface chemistry of a specific adhesive may be such that a strong stickie particle/air bubble complex is formed, with its subsequent removal.

Some laboratory studies suggest stickies can be removed by flotation. In one of these, three hot melts were removed very effectively during laboratory flotation provided the attachment was above a minimum attachment force. These experiments were under controlled conditions, in the absence of fiber. Hot melt particle sizes were given as 57-420 m and air bubble sizes of 260 m and 350 m. As removal efficiency was reported as high as 95%, this data implies the removal of stickies that were larger than the air bubble. This is very different from mineral flotation, where the optimum air bubble to particle size ratio was estimated to be about 5:1. Although this study showed stickies removal was possible, again the conditions used were very different from those in a mill.

Care needs to be taken when comparing laboratory deinking results with industrial scale

In another study using three pressure sensitive adhesives, stickie removal rates by flotation were 30-35%. The stickie particle size range was 320-2,500 m, well above the size range of ink particles that would be expected to be removed by flotation. The authors remarked on the tenacity of the attachment of the stickie particles to the air bubbles. In only one case (styrene/butadiene) was the removal efficiency affected by the size of the stickie particle.

Equally confusing were the results from a study in a German newsprint mill. In this case, no removal of macro stickies was found in the pre-flotation cells. Laboratory tests showed that stickies removal was theoretically possible under different conditions.

Removal was also evaluated in a flotation stage after high-speed dispersion, the post-flotation step. Following a plant upgrade, the stickie removal efficiency in the secondary flotation stage (of the post-flotation step) was more than 80%, whereas in the primary stage it was about 56%. Overall removal across the two stages of the post-flotation step was thus about 45%. Macro stickies were defined as those in the size range 150-500 m.

The authors of this study did not speculate on the reasons for the difference in stickie removal efficiency between the two flotation steps. Usually their chemistry is quite different and it may be that deinking chemistry used in the pre-flotation step was incompatible with stickies removal. Changes to the chemistry in the post-flotation step (or changes caused by dispersion) resulted in the ability of the post flotation system to remove stickies.

Future flotation

Much about the ability of flotation to remove stickies in mill situations remains to be discovered. This may well lead to changes in flotation cell design and chemistry.

Roland McKinney is an independent consultant specializing in recovered paper studies such as its recovery and all aspects of processing. His company, Fiber Research Consultants is based in the UK, and he can be reached on tel +44.1372.458.335 or by fax on +44.1372.458.335. If you have any comments on flotation he would be pleased to hear from you