Sizing Problems (Acidic, AKD, ASA, Rosin, Self-sizing, Size reversion, Variable sizing)

The purpose of internal sizing is to resist the penetration or spreading of liquids through or on paper. The subject is complicated by the fact that there are several very different kinds of chemical treatments to make paper fluid-resistant, there are widely different kinds of liquids with which the paper may interact, and sizing is affected by many different variables at the wet end of the paper machine. For a more detailed description of various aspect of sizing technology, the reader is encouraged to look up some of the references at the end of this section. For simplicity, we will assume that the fluid of interest is water or an aqueous solution.

The primary wet-end additives for decreasing the penetration or spreading of aqueous solutions through or across paper are known as internal sizing agents. Some separate comments will be provided in later paragraphs regarding three major product categories of internal sizing agents, alkenylsuccinic anhydride (ASA), alkylketene dimes (AKD, and rosin. But to begin with, it can be useful to keep in mind the following key factors upon which all of these products depend for their efficient use:
1. The sizing formations, before they are added to the system, need to be in a well dispersed form.
2. The formulations need to become well mixed with the fiber furnish so that the material can be relatively uniformly distributed onto the solid surfaces.
3. The sizing agents, together with any fines to which they are attached, need to be retained in the paper at the forming section. Retention efficiency is particularly important in those cases where sizing agents gradually become converted to a less active or inactive form during exposure to the furnish. This is especially important in the case of ASA size, but it also can be important for the performance of AKD or in the case of rosin acid emulsion sizing products when the pH is higher than about 6.
4. For the most efficient sizing, individual molecules of sizing agent need to be distributed over the surfaces of solids in the paper. Though there may be a limited ability of some sizing agents to spread in liquid form over the surfaces to which they are attached, evidence suggests that vapor phase transport during the drying of paper is critical for sizing with ASA, AKD, and emulsion-type rosin products.
5. The sizing agent molecules need to be anchored and oriented on the fiber surface. In the case of ASA this appears to involve a curing reaction in which the anhydride reacts with hydroxyl groups on the fiber surface to form an ester. In the case of AKD there has been some dispute about the mechanism. The ketene dimer structure appears to be well suited for distribution of the product over the paper surface, but it is not clear how much of it is able to form ester bonds. Rosin products are most effective when they are able to form insoluble aluminum compounds. In the case of rosin soap products, these compounds are formed near to the point of addition of the rosin or alum. In the case of rosin acid emulsion products, the aluminum rosinate compounds mainly are formed during the drying of the paper.


Some factors that can cause inadequate resistance to penetration or spreading of aqueous liquids through and across paper include (a) the presence of large pores due to low paper density or the presence of coarse fibers, (b) pin-holes, possibly due to entrained air, (c) an inadequate dosage of sizing agent, (d) premature decomposition of the sizing agent, especially if it is stored too long or if it remains in contact with hot furnish too long because of low retention efficiency, (e) the presence of surface-active materials, as in the case of nonionic surfactants used in deinking of paper, (f) inadequate drying of the paper to achieve full cure, especially in the case of AKD size, (g) unexpectedly high level or surface area of minerals, especially in the case of precipitated calcium carbonate products, (h) inadequate amount of soluble aluminum additives, if one is using a rosin-based sizing agent (should be at a ratio of about 1.5:1 alum on the mass of size, in addition to the amount of alum needed for such purposes as charge neutralization of the furnish), (i) an inadequate amount of cationic starch (should be at a ratio of about 3:1 on sizing agent) during formulation of the ASA emulsion.

Some strategies to increase the effectiveness of liquid hold-out parallel the list just given. To reduce the average pore size in a sheet of paper made from kraft fiber, it makes sense to increase the level of refining. Though pin-hole problems often can be solved by increasing the use of a defoamer, it is important to proceed with caution. Defoamers usually contain surface-active materials, and these tend to work against the hydrophobic action of internal sizing agents. Fortunately, there are many defoamer products that have minimal impact on internal sizing.

In the case of ASA sizing, some special strategies can be used to make the process work as efficiently as possible. Strategies to minimize the extent of decomposition of the sizing agent include (a) delivery of the emulsion to the process immediately after it is formed, without using any holding tank, (b) cooling the solution of cationic starch before it is used to emulsify the ASA, (c) buffering the cationic starch solution into the acidic range by adding alum or adipic acid, (d) adding the product relatively late in the process, usually after the hydrocyclone cleaner system, and (e) employing an effective retention aid system. The addition of at least a small amount of soluble aluminum product to the system has been shown to improve the efficiency of ASA sizing and to minimize ASA hydrolyzate deposit problems.

In the case of AKD the most critical issues usually relate to retention and curing. The lower reactivity of AKD, compared to ASA, makes it feasible for suppliers to deliver AKD in a ready-to-use formulation. It also is feasible to add AKD as far back in the system as the outlet from the machine chest, into the stuff box, or at the inlet to the fan pump. Sometimes these strategies help to retain a larger proportion of the sizing agent onto the surfaces of long fibers, for which the first-pass retention is near to 100%. Although the curing of AKD usually is favored by the presence of carbonate alkalinity, it does not follow that sizing is favored by high levels of calcium carbonate mineral. Rather, it has been found that association of AKD onto calcium carbonate surfaces tends to yield temporary sizing, which can revert over time. The effect seems to be more significant when using precipitated calcium carbonate products, which tend to have higher surface areas and higher pH values than ground calcium carbonate products. Separating the addition points of the filler and the sizing agents, either by time or by the addition of cationic starch, may help to minimize this kind of problem. Some AKD size products contain cationic resins that seem to act as cure promoters or size retention promoters.

In the case of rosin emulsion products it is important to have sufficient soluble aluminum additive to the system so that the size will have something to cure onto when the molecules are vaporized and recondensed in the dryer section. The alum can do double-duty as a retention aid for the rosin. It is most common to use a "reverse" addition order, in which the alum is added first. The elapsed time between addition of the aluminum product and the sizing agent ought to be as short as practical, except that the first additive should be well mixed with the furnish before the second is added. Such an order of addition tends to favor the formation of rosin-aluminum compounds, rather than the undesired formation of rosin-calcium salts in the case of rosin molecules that are present in their carboxylate form.

The ideal conditions for curing of rosin emulsion products is generally in the pH region that favors formation of oligomeric, highly cationic and flocculated forms of alum or poly-aluminum chloride (PAC), i.e. 4.5 < pH < 6. However, conditions of high temperature and pH will tend to saponify the rosin prematurely, which is not desirable. In order to "push" rosin system to work at pH values above 6, papermakers use such strategies as (a) careful premixing the emulsion size with PAC, an approach that would not be wise at lower pH values, (b) adding the formulation very late in the process and aiming for very high retention, and (c) using specialized rosin products; for instance some rosin esters may be used in place of some of the rosin acid in the formulation.

Rosin soap sizing generally works best at rather low pH values of about 4 to 4.5, or perhaps as high as 5. If the process water is relatively soft (calcium ion content below 50), it has been usual practice to add the rosin before the alum, the so-called "forward" order of addition. However, if the water is relatively hard, it is still better to add the alum ahead of the rosin soap size to minimize the formation of calcium rosinate. Good agitation after the point of addition of the rosin will help form very small particles of aluminum rosinate, which can help make the process more efficient.

If there is already a lot of size product being put into the product, and there doesn't seem to be anything the matter with any of the additive systems, try to find out what is happening to what has been added. Are there any accumulations of material that might be sizing agent in the wet presses, driers hoods, or in deposits elsewhere in the system? Is there more foam than usual? Such observations may point to a root cause.


Factors that are likely to cause variability of sizing are some of the same factors that can cause shifts in retention efficiency and other changes in the wet end. One needs to find out whether changes in water resistance of the paper or in the demand for sizing agent can be correlated to other cycles in the system. For instance, there may be episodes of increased levels of surface-active materials that act as anti-sizing agents. Such materials may include black liquor carry-over, components of coated broke formulations, nonionic surfactants from deinking operations, components of dye formulations or slimicide formulations, and various anti-foam surfactants. The amounts or the surface area of filler may be cycling up and down, or maybe there are instabilities in the retention aid system. It is also possible that variations in the cationic demand of the furnish are causing cycles in retention efficiency, which are also affecting sizing efficiency. It may help to review some of the key factors mentioned in the previous subsection about different classes of sizing agent.


Size reversion can be defined as a decrease in fluid resistance over time, as paper is stored or shipped. The effect can be observed occasionally in most kinds of sized paper, but it is most prominently noted in alkaline paper with AKD or ASA sizing agents. This makes sense, since these systems generally contain less sizing agents to begin with, and there tends to be a strong threshold effect. In other words, no sizing effect is observed at dosages up to about 1.5 lb/ton, depending on the furnish, and thereafter the sizing effect increases sharply. It makes sense that changes over time can have the effect of shifting the threshold level to higher values of size treatment. As noted earlier in this section, reversion of AKD sizing often can be reduced by minimizing the contact between sizing molecules and calcium carbonate. Strategies that can be used to achieve this include (a) greater separation of the addition points for AKD formulation and calcium carbonate filler, (b) reducing the surface area of the filler or switching to ground calcium carbonates, especially chalk, or (c) reducing the filler level.

Rosin sizing may be unstable in some cases if there is insufficient interaction with alum. This situation is most likely to be encountered when using rosin emulsion products. Though the emulsion products have the advantage of retaining their fluidity until the point of cure in the dryers, they have the disadvantage that there is very little time for the molecules to encounter and interact with precipitated aluminum compounds on the fiber surfaces. Such a situation can be improved by practices that ensure an even distribution of finely divided rosin emulsion particles over the fiber surfaces, in the presence of an adequate amount of alum or PAC treatment. If the paper is stored in a hot roll, the size appears to partially evaporate from the hot areas and recondense in the cooler zones of the paper.


The term self-sizing has been most often used to describe an effect that happens in relatively high-yield furnishes, such as newsprint or corrugating medium. The paper or board may have a low ability to hold out aqueous liquids when it is first made. However, the paper becomes more hydrophobic during storage, especially at high temperatures. The effect has been attributed to migration of rosin acid molecules that were naturally present in the softwood fibers in the furnish. Papermakers sometimes add surface-active agents (wetting agents) to overcome self-sizing effects in cases where this factor is important. Also it may be important to eliminate soluble aluminum additives when producing products such as paper towels, which are intended to be highly absorbent.

More ideas about optimizing of the degree of hydrophobic size treatment can be found in an essay that is titled "Right Sizing."


Bartz, W. J., Darroch, M. E., and Kurrle, F. L., "Alkylketene Dimer Sizing Efficiency and Reversion in Calcium Carbonate Filled Papers," Tappi J. 77 (12): 139 (1994).

Bottorff, K., "AKD Sizing Mechanism: a More Definitive Description," Tappi J. 77 (4): 105 (1994).

Chen, G. C. I., and Woodward, T. W., "Optimization of Alkenyl Succinic Anhydride Emulsification and Sizing," Proc. TAPPI 1986 Papermakers Conf., 37 1986).

Colasurdo, A. R., and Thorn, I., "The Interactions of Alkylketene Dimer with Other Wet-End Additives," Tappi J. 75 (9): 143 (1992).

Hodgson, K. T., "A Review of Paper Sizing using Alkylketene Dimer versus Alkenylsuccinic Anhydride," Appita 47 (5): 40 (1994).

Hoyland, R. W., and Neill, M. P., "Factors Affecting the Frictional Properties of Paper - the Effect of AKD Neutral Size," Paper Technol. 42 (3): 45 (2001).

Hubbe, M. A., "Wetting and Penetration of Liquids into Paper," in Encyclopedia of Materials Sci. Technol., Elsevier, Oxford, 2001.

Ito, K., Isogai, A., and Onabe, F., "Rosin-Ester Sizing for Alkaline Papermaking," J. Pulp Paper Sci. 25 (6): 222 (1999).

Kitaoka, T., Isogai, A., and Onabe, F., "Sizing Mechanism of Emulsion Rosin Size-Alum Systems," Nordic Pulp Paper Res. J. 12 (1): 26 (1997).

Liu, J., "Sizing with Rosin and Alum at Neutral pH," Paper Technol. 34 (8): 20 (1993).

Marton, J., "Practical Aspects of Alkaline Sizing: Alkylketene Dimer in Mill Furnishes," Tappi J. 74 (8): 187 (1991).

Marton, J., "Mechanistic Differences between Acid and Soap Sizing," Nordic Pulp Paper Res. J. 4 (2): 77 (1989).

Moyers, B. M., "Diagnostic Sizing Loss Problem Solving in Alkaline Systems," Proc. TAPPI 1991 Papermakers Conf., 425 (1991).

Ness, J., and Hodgson, K. T., "The Effects of Peroxide Bleaching on Thermo-Mechanical Pulp Self-Sizing," Nordic Pulp Paper Res. J. 14 (2): 111 (1999).

Patton, P. A., "On the Mechanisms of AKD Sizing and Size Reversion," Proc. TAPPI 1991 Papermakers Conf., 415 (1991).

Petander, L., Ahlskog, T., and Juppo, A. J., "Strategies to Reduce AKD Deposits on Paper Machines," Paperi Puu 80 (2): 100 (1998).

Savolainen, R. M., "The Effects of Temperature, pH, and Alkalinity on ASA Sizing in Alkaline Papermaking," Proc. TAPPI 1996 Papermakers Conf., 289 (1996).

Strazdins, E., "Paper Sizes and Sizing," in D. F. Zinkel and J. Russell, Eds., Naval Stores, Pulp Chemicals Assoc., New York, 1989, Ch. 16, p. 575.

PLEASE NOTE: The information in this Guide is provided as a public service by Dr. Martin A. Hubbe of the Department of Wood and Paper Science at North Carolina State University ( Users of the information contained on these pages assume complete responsibility to make sure that their practices are safe and do not infringe upon an existing patent. There has been no attempt here to give full safety instructions or to make note of all relevant patents governing the use of additives. Please send corrections if you find errors or points that need better clarification. Go to top of this page.

Home page Troubleshooting EncyclopediaEducational opportunities Research opportunities Business opportunities What's new in the field? Background information Links to wet-end chemistry Fun Stuff E-Mail