Volume 21, Number 1
May 1997

Scientific Studies on Best Conditions for Long-Term Storage

This is a condensed report of research done for the Commission of the German Library Institute for Maintaining Holdings, by Battelle Ingenieurtechnik GmbH, Eschborn. It is the Final Report for Stage II, dated May 1994. Helmut Bansa's name is on the German-language title page as "Koreferat" (consultant?). Translation was done by the Institute for Paper Science and Technology in Atlanta.

For a German-language copy of the complete report, contact Ulrich Behrens, Battelle Ingenieurtechnik GmbH, Düsseldorfer Strasse 9, 65760 Eschborn, Germany (tel: 49/6196/936-0; fax 49/6196/936-499). Ask for "Wissenschaftliche Arbeiten zur Ermittlung der optimalen Bedingungen für die Langzeitlagerung von archivalischem und bibliothekarischem Sammelgut."

Endogenous and exogenous influences subject paper to a natural degradation process. In the case of industrially manufactured papers, which make up the major portion of the book and document holdings in libraries and archives, this degradation process is primarily caused by the sulfuric acid contained in the paper as well as by the organic acids forming during decomposition of materials accompanying the cellulose (especially lignin).

Chemistry teaches us that the rate of chemical processes is reduced by lowering the temperature. In order to retard the disintegration of books and documents in libraries and archives, it is necessary to introduce so-called "cold storage" (i.e., preservation of books and documents at low temperatures and humidities) into the storage areas. Considered as optimum are temperatures near or even below the freezing point. The climatic conditions under which a physically inactive mentally occupied person is comfortable differ clearly from those which are needed for optimal preservation in the storage area. They lie at 20-30°C and 40-60% RH. The climate in reading rooms must therefore be set in this temperature and humidity range since the users of libraries and archives stay and work practically exclusively in these rooms and often over longer periods of time.

If "cold-stored" books or documents are brought from the storage area into the reading room and then back from the reading room into the "cold-storage area," the environment for the paper changes twice within a relatively short time from a higher to a lower temperature and back again, with a corresponding change from a lower to a higher relative humidity.

Paper takes up moisture from the surrounding air when it is brought out of a cool and dry climate into a warm and moist one. Conversely, one should count on a release of moisture from the paper upon transfer from a warm, moist climate into a colder, drier one. This release and uptake of moisture causes a dimensional change in the paper (changes in volume for the cellulose fibers and the void spaces between the fibers).

The goal of the overall project is to determine the influence of frequent climatic-change stress on the loss of strength in papers. The aim was to determine experimentally which climatic conditions permit the coldest and driest possible storage, on the one hand, and the use of the object in a climate comfortable for the user, on the other.

In addition, one needs to study how to select a storage climate that can be achieved in buildings as easily as possible and without great technical effort.

The theoretical and experimental data determined in Stage I of the overall project showed that the uptake and release of water by the paper is determined by the temperature and the relative humidity. It was the goal of the first phase in Stage II to determine experimentally the climatic range, i.e., the temperature and humidity parameters at which the uptake and release of water is at a minimum when these parameters change. At the same time, the upper range of this climate should be adjustable such that it is perceived as "comfortable" by users of the reading room.

It was the goal of the second phase of Stage II to investigate via a long-term experiment the effects on paper strength of climatic-change stress during "optimum" and "non-optimum" conditions in storage and reading rooms.

The experiments were carried out with nine selected paper grades forming a representative cross-section of papers most frequently used in library books and in documents in German archives. These included six "new, modern" papers and three "historic" papers. The "historic" papers were a non-deacidified and a deacidified paper as well as a rag paper.

The testing was done in two commercial aging chambers, which were run in the circulated-air mode for the experiments. Located in the path of the circulating air were the necessary cooling and heating elements, the fan unit, and the humidifying and dehumidifying system. The air was conditioned according to target values and circulated through the chambers. The aging chambers were tight closed systems.

The amount of water absorbed and released by the papers as climatic conditions changed was determined by weighing the individual paper sheets. For this purpose, 150 sheets of each of the new papers and 100 sheets of each of the historic papers were fastened together as a unit. The papers were suspended vertically on three rods with a spacing of about 3 mm between the individual sheets. Weighing was carried out continuously by means of weighing cells.

The experimental results from the first phase showed that, within the framework of measuring accuracy, little or no uptake and release of water takes place when the temperature changes but the relative humidity is held constant. This finding is valid for the investigated temperature range between 7°C and 29°C and for the humidity range between 20% RH and 80% RH.

The long-term investigations of the second phase were based on the results of the first phase, using the following climatic parameters:

Nonoptimal Climate C I: Climate with significant uptake and release of water

Storage climate: 7°C/30% RH

Reading-room climate: 29°C/80% RH

Optimal Climate C II: Climate with low uptake and release of water

Storage climate: 7°C/50% RH

Reading-room climate: 29°C/50% RH

The simulation of an optimal climate was carried out at a relative humidity of 50% RH. This RH value lies in the middle range of the investigated humidities. However, every other RH value can be viewed as optimal as long as it is identical for Climate I and Climate II.

The long-term investigations concerning the influence of climatic-change stress were carried out in parallel in two aging chambers using the above-mentioned nine paper grades. In these studies, the individual climatic values were held constant over four hours so that sufficient time was available for the establishment of equilibrium moisture in the papers. The establishment of equilibrium is shown by constancy of weight as a function of time. The transition from Climate I to Climate II and from Climate II to Climate I was realized in two hours so that four climatic changes were simulated in 24 hours. The experiments began in July 1992 and ended in December 1993. Overall, the papers were subjected to about 1450 climatic changes.

At regular intervals, paper samples were taken and investigated at the Institut für Papierfabrikation (Institute for Paper Manufacture) at the Technische Hochschule Darmstadt (Darmstadt Technical College) for changes in strength. The studies showed that the strength performance of the papers exhibits the same tendencies under the nonoptimal climate C I and the optimal climate C II and is to be viewed as equivalent within the framework of measuring accuracy.

On the basis of the test results and a working hypothesis derived therefrom, the following can be said:

• Within the test period and under the test conditions, one is unable to recognize any influence of uptake or release of moisture by itself on the strength and thus on the aging of paper.
• Climatic changes allow a tendency toward an increase in aging to be recognized. Since moisture can be ruled out as a significant influence, this is traced to the thermally induced mechanical stressing of the papers as a result of temperature changes. The six unstressed modern papers showed a tendency toward an increase in breaking load after 1452 climatic changes between ca. 3% and 20% for climate C I and between 3% and 16% for climate C II. The increase in breaking load is to be traced to new bonds or to a bond-reinforcing reorientation of existing bonds between the cellulose fibers, caused by decomposition of hemicelluloses and amorphous fiber regions. This leads to an increase in brittleness. In line with this, one finds in the case of most of the modern papers investigated a clear decreasing tendency in elongation at break of about 2.5% to 20% with climate C I and about 5.8% to 13% with climate C II. Exceptions are Papers 2 (Haindl) and 3 (Scheufelen), for which a minimum increasing tendency in elongation at break of 1% to 2% is obtained from the theoretical regression for Climate C I. Under Climate C II, however, these papers show a different behavior for the theoretical regression. While Paper 2 (Haindl) likewise exhibits a decreasing tendency of about 6%, one obtains for Paper 3 (Scheufelen) an increase of about 4% even under C II. However, an analysis of the measured values for Papers 2 and 3 revealed that, in the end, a decrease in the elongation at break should be assigned to these papers too, since, in contrast to the other papers, the theoretical regression curves are so strongly affected here by individual extreme values or by an initial increase in values that unclear or false tendencies may have been recorded.
• The decrease in MIT double-fold number under the same limit conditions amounts to 28% to 58% for Climate C I and to 37% to 57% for Climate C II.

Since an increased storage temperature accelerated chemical reactions in the paper, including both "normal" degradation and that related to "aging" as well as damage by anthropogenic airborne materials and by the constituents of air, it is necessary to store at lower temperatures than those involved in use. The test results presented here show that temperature changes will stress the papers and apparently cause aging, as demonstrated by the recognizable tendency toward a decrease in strength. If a reduction in strength due to temperature change is to be excluded, the difference in the temperature of the storage and use rooms must be kept as low as possible.

According to the present state of knowledge, the storage of papers and books at low temperatures retards aging. Since, at the same time, a decrease in strength is found to result from temperature changes, one needs the most quantitative information possible concerning the extent of aging caused by temperature changes in order to establish optimum climatic conditions for storage and use rooms.

Cold storage is sensible if the strength reduction caused by temperature-change stress is considerably less than preservation of strength resulting from cool storage. This means that the papers have experienced a strength loss, to be sure, as a result of the different temperatures of the storage and use rooms over several decades, a strength loss which, however, is more than compensated for by cool storage so that the paper strength resulting from the cool storage is still far above that of papers stored under the currently conventional storage conditions of 18°C and 50%RH.

However, estimating tolerable temperature differences between storage and use rooms appears to be possible only if the strength loss of the papers investigated here after artificial aging over various time periods can be related to the strength loss caused by temperature changes.

It is not possible to freely transfer the results found in these experimental investigations on unbound and unstressed modern papers to bound papers provided with covers of the widest range of individual or composite materials.

Based on the results obtained, the studies to be carried out in the planned Stage III of the overall project, which are aimed at developing practical designs, should first be begun only after reliable data have become available on the different expansion coefficients for the materials making up a book and on the resulting physical-mechanical factors depending on the frequency and extent of humidity and temperature changes. Only then will it be possible to give the fluctuation range of climatic values within which mechanical damage does not occur. This finding is important since maintaining the climate (especially the relative humidity) constant in a reading room frequented by the public can be realized technically only with a complicated system and high costs.

Additional studies required for a comprehensive report on the behavior of unbound records and bound papers were identified.