JAIC 1987, Volume 26, Number 1, Article 1 (pp. 1 to 17)
JAIC online
Journal of the American Institute for Conservation
JAIC 1987, Volume 26, Number 1, Article 1 (pp. 1 to 17)


Antoinette Dwan


THE RATE AND AMOUNT of moisture regain is dependent on time and the chemical source of the cellulose. The more crystalline the paper, the lower the capacity for absorption. Moisture uptake is greater in papers with higher hemicellulose or lignin content. It is also greater in papers altered by chemical treatments which reduce crystallinity,20 such as oxidative bleaching.

Moisture content depends not only on chemical constituents and physical properties, but also on the manner in which the paper was brought into equilibrium. There is a different moisture content at any given relative humidity depending on whether the moisture was desorbed (brought into equilibrium from a higher relative humidity) or adsorbed (brought into equilibrium from a lower relative humidity). The curves for desorption and adsorption are not the same. This is known as the hysteresis effect, and is shown in Figure 11.21

Fig. 11. Hysteresis Curve. “A” is Desorption Curve. “B” is Adsorption Curve. (Reprinted, with permission, from TAPPI, June 1961)

Figures 12, 13, and 14 show hysteresis curves for samples of three types of paper that were properly preconditioned. Each of these figures demonstrates that the data obtained is different at any relative humidity depending on whether the measurement was made after desorption or adsorption. Conservation studies that compare before and after aqueous treatments of any type (humidification, steaming, bathing) are analogous to this situation. Interpretation of test results must account for the magnitude of change in results due to hysteresis effects.

Crook and Bennett demonstrated the effect of hysteresis on fourteen papers through numerous tests. They state:

In general, especially for the weaker papers, the desorption curve was associated with the higher folding endurance at most humidities. In some instances, however, the adsorption and desorption curves crossed.

The total change in strength, over the relative humidity range investigated, of the papers other than the machine-glazed papers and index card, varied from 50% in the case of bible tissue and litho to 250% in the case of chart paper.22

Prior history of moisture gain and loss also affects moisture regain capacity and paper properties. Since paper is a viscoelastic material, it does not return to the same dimensions after drying and rewetting, or after humidity cycling. There is a progressive change in dimension after each humidity cycle, with the greatest change occurring after the initial cycle.23(Figure 15)

Fig. 15. Effect of Repeated Humidity Cycling on the Dimension of Paper, the Machine Direction. (Reprinted, with permission, from TAPPI, June 1961)

Wink studied dimensional changes in linen writing paper during changing relative humidity conditions. He exposed samples to the humidity cycle of 50%, 11%, 50%, 94%, 11% in increments of about 15% R.H. He repeated the cycle eight times, with the temperature at 73F. Wink found appreciable shrinkage at the three relative humidity levels. The curves show that the stress-relaxation effect dominates the swelling effect to such a degree that all dimensions at 94% relative humidity were smaller than the initial dimensions at 50% relative humidity.24

Irreversible effects, resulting from an excursion of paper to a high relative humidity, are often observed. These can be of an appreciable order of magnitude, with the properties, in certain cases, altered to such an extent that they no longer characterize the original material. This effect evidently originates with the swelling and shrinking of the fibers and with the relaxation of dried-in or built-in stresses; the major effect occurs on the first exposure of paper to a high relative humidity, exceeding approximately 65%; it is dependent upon the extent of the excursion and it permanently alters such surface properties as gloss and smoothness, as well as dimensional and strength properties. These changes are nonrecoverable by manipulation of the moisture content or by preconditioning and conditioning the paper.25

Moisture is so critical to the properties of paper that the TAPPI standard for temperatures and moisture require a temperature of 23 1C and a relative humidity of 50% 2%.26 The TAPPI standard acknowledges the discrepancy caused by the hysteresis effect in moisture content and recommends preconditioning the sample from below 50%, by absorption. This same standard emphasizes the effect of preconditioning and the hysteresis effect in the section, “Importance of Preconditioning”:

The physical properties of a sample at 50% R.H. depend on whether the sample was brought to 50% from a higher or lower R. H. This humidity hysteresis effect is a 5–25% of the test value for many physical properties. Conditioning down to 50% gives most papers a moisture content very nearly the same as conditioning up to 60%.

While preconditioning procedures practically eliminate the hysteresis effect, it has little influence on strain relaxation effects. The latter depend on the entire previous moisture history of the sample, especially on the conditions of initial drying and tension, and on the duration and degrees of subsequent excursions to high humidities (i.e. above about 58% R.H.)27.

Crook and Bennett emphasize the importance of the hysteresis effect when they conclude:

Appreciable hysteresis differences are observed between test values obtained by conditioning from a low humidity to a higher humidity and those obtained when the initial conditioning has taken place at a high humidity. In some instances the subjection of samples to very high humidities brings about a permanent alteration of the strength of the paper.28

Researchers often use handsheets prepared in the laboratory under controlled temperature and relative humidity conditions. The use of handsheets eliminates effects caused by the unknown history of humidity cycling, hysteresis, moisture content, creep, and stress-relaxation. The use of papers other than handsheets (for instance historical papers) presents unknown variables making correct interpretation of data complex or impossible.

Copyright 1987 American Institute for Conservation of Historic and Artistic Works