FACTS GENERAL INFORMATION ARTICLE FACTS publishes this document as a public service. Its use is voluntary, and all results obtained by its use must be entirely the responsibility of the user. This document is subject to revision, change and/or withdrawal at any time.  © FACTS 2000

Book ageing

N. Gurnagul, R. C. Howard, X. Zou, T. Uesaka and D. H. Page

THE MECHANICAL P]

A LITERATURE REVIEW

CE OF PAPER:

ABSTRACT: Librarians and archivists are concerned about book pages that become brittle and disintegrate as a consequence of ageing. This has resulted in an increased demand for paper with good mechanical permanence, i.e. paper that retains its strength properties with ageing. This report reviews the literature on the factors that affect the mechanical permanence of paper.

INTRODUCTION

Canadian producers of printing, writing and publication grades are coming under increasing pressure to manufacture neutral or alkaline paper. This pressure has arisen from concern about the permanence, or lasting qualities, of printed documents. Such concerns have been reflected in a recent government decision to print publications it expects to be retained for information or historical purposes on alkaline-based permanent paper [1].

A Canadian standard for permanent paper is not yet available. For the time being paper for government publications must satisfy the 1984 version of the American National Standards Institute (ANSI) specifications for permanence [2]< This standard specifies that the paper must have a minimum cold extract pH of 7.5, an alkaline reserve equivalent to 2% calcium carbonate, must not contain any groundwood or unbleached pulp, should have an initial folding endurance of 30 double-folds (at 1 kg tension) in the cross-direction and a tear resistance in the machine-direction of 24-50 grams (for 30-60 lb paper). This standard has recently been revised as follows: paper stock requirement has been changed to less than 1% lignin in place of the 1984 specification of no groundwood or unbleached pulp; in addition, the folding endurance test has been dropped due to the inherent variability of the test [3].

l First published in the Journal of Pulp and Paper Science, 19 (4), J-160­166, 1993 and reprinted here with the Journals's permission.

100°C, humid air at 95°C, exposure under an arc lamp at unspecified temperature and humidity, and exposure to steam at atmospheric pressure [15]. Testing a wide range of commercial papers, Rasch concluded that dry air at 100°C produced the most severe ageing effects, and the most reproducible. He also proposed that the folding endurance after 72 hours at 100°C, expressed as a percentage of the initial value, should be termed the "permanency factor". The method was subsequently adopted by NBS and was used by numerous workers [16-21].

Relating artificial ageing to natural ageing is certain to be subject to error, when the circumstances of natural ageing can be so varied. By 1962, Neimo t6] was able to refer to 58 different laboratory ageing tests which employed conditions ranging from a duration of 1 hour to 180 days, at relative humidities from 1-100%, and temperatures from 22-130°C. It was recognised at that time that by ageing in dry air, the potential importance of sheet moisture was being overlooked. Thus, Browning and Wink t22] used 5 temperatures between 60 and 120°C, at 2 levels of relative humidity (RX). They found that increased humidity corresponded to faster degradation of sheet physical properties; other workers came to similar conclusions t23-25].

The underlying assumption of an accelerated ageing test is that it simply speeds up the chemical reactions which would normally occur under natural ageing conditions. This is true if the same chemical reactions occur under both natural and accelerated ageing and that the wrelative" rates of these reactions are also similar. Recent work by Arney and Jacobs suggests that this assumption may be incorrect t26]. They show that the importance of atmospheAc oxidation to the overall rate of deterioration of newsprint decreases as the testing temperature is lowered. This suggests that accelerated ageing at high temperatures may lead to the wrong conclusions regarding the chemical reactions that may be occurring during the natural ageing process. There is clearly a need for a detailed kinetic analysis to determine the rate constants and activation energies for each possible chemical reaction that may be occurring during accelerated ageing; this will allow for more meaningful comparisons with natural ageing.

Based on the above, it is not surprising that the scientific community has yet to reach a consensus on what conditions are appropriate for an accelerated ageing test. Currently Tappi has two provisional standards for estimating the relative stability of paper: dry heating at 105°C [27] or moist heating at 90°C and 25% RH [28] for different time intervals. On the other hand ISO suggests ageing at 80°C and 65 % RH [29]. Regardless of the controversy surrounding accelerated ageing tests, several workers have compared natural ageing with accelerated ageing and have concluded that accelerated ageing is suitable at least for ranking the stability of a group of papers relative to one another rls.2l*30,3l].

Standards on permanence are also being issued by other organisations such as the International Standards Organization (ISO) [4] and the American Society for Testing and Materials (ASTM) [5]. Like the ANSI standard these have similar requirements with regard to lignin content, pH and ~line reserve.

A number of good reviews exist on paper permanence [6-11]; however these do not necessarily focus on the issues raised in the standards for permanence such as the effect of pulp type. This review covers the literature from 1895 to the present with an aim to reveal the anecdotal and scientific basis for the specifications outlined in the various standards for paper permanence.

DEFINITION OF P]

Permanence was defined in the early 1920's as "the degree to which paper resists chemical action which may result from impurities in the paper itself, or agents from the surrounding air" [12]. This lead to the emphasis on high fibre "purity", i.e. a high a-cellulose content as one of the requirements for permanence [13].

This definition of permanence does not however take into consideration the end-use requirements for the paper. Though chemical changes may occur within the sheet during ageing, they will not necessarily lead to changes in the end-use performance. The basic requirement from the point of view of the archivist is to be able to handle the pages of a book without the sheets crumbling into pieces. Thus the retention of the sheet's mechanical properties should be the most important requirement for archival materials. This study will therefore be confined to the consideration of mechanical permanence.

ACCELERATED AGEING TESTS

Since the processes of ageing under ambient conditions are extremely slow the only approach to a scientific study of ageing is some form of an accelerated ageing test. The question is how best to accelerate the processes occurring during natural ageing, without inducing artificial effects.

The basis for accelerated ageing tests is the well-known principle that chemical reactions are faster the higher the temperature. Therefore it is not surprising that dry heat treatment was proposed to promote ageing in paper as early as 1899 [14]. In 1926, Gosta Hall compared the permanence of rag and sulphite handsheets following exposure to heat and sunlight [12]. In 1929, Rasch, at the National Bureau of Standards (NBS), used 4 methods: dry air at

The prediction of paper permanence under natural ageing from a knowledge of the activation energy thus becomes somewhat uncertain although it continues to be a valuable research goal.

CHEMICAL REACTIONS OCCISRREG DUR1NG AGEING

The chemical reactions occurring during natural ageing are not completely clear. Based on the chemical and physical analysis of natural and accelerated aged samples, several reactions have been proposed that may be linked to the loss of mechanical properties; these are: 1) acid catalysed hydrolysis, 2) oxidation and 3) cross-linking t39]. It is possible that all these reactions occur, but at different rates in different circumstances or for different papers.

Hydrolysis

A significant amount of data has been gathered on the cold extract pH of old documents suggesting that acid hydrolysis is the most important cause of paper embrittlement t40]. The sources of acidity in the sheet have been linked to acidity of the papermaking process, particularly resulting from the use of alum during the rosin sizing process, as well as atmospheric contamination and other miscellaneous causes such as writing inks.

Alum (A12 (S04)3 . 18H2O) has frequently been identified as the primary source of residual acidity in the sheet but the mechanism by which alum degrades cellulose is less clear. The conventional view is that alum hydrolyzes in the presence of water to produce acid which then depolymerizes the cellulose t41,42]. This view has been challenged by Parks who noted that differential thermal analysis (DTA) plots of cellulose treated with alum are distinctly different from a cellulose sample treated with sulphuric acid [43], suggesting that acid produced from the reaction of alum and water is not the only factor in the degradation of cellulose.

Early studies have shown that papers exposed to atmospheres containing sulphur dioxide become brittle t44-4Q- Subsequent studies by Langwell have shown that metallic impurities such as iron and copper in the sheet catalyse the oxidation of atmospheric sulphur dioxide to sulphur trioxide (S03), and thus to sulphuric acid [47-52]. Hudson and co-workers used radioactive sulphur dioxide as a tracer to confirm Langwell's findings and showed that mechanical pulps have a higher affinity for sulphur dioxide than rag pulps t53-55]. They speculated that this was due to the presence of ketone and aldehyde groups in the mechanical pulp furnish; however they did not consider the effect of sheet structure. Mechanical pulp sheets are more porous and have a higher specific surface area due to the presence of fines, thus one may expect increased sorption of atmospheric gases based on sheet structure considerations. Hudson's data focuses primarily on the amount of sulphur pickup by different papers, it

CE UNDER NATURAL AGEING

PREDICTION OF PAPER P]

The concepts of chemical kinetics are often applied to accelerated ageing experiments to estimate how long paper will last under natural ageing conditions. For a chemical reaction Arrhenius has shown that the experimentally observed variation of the rate constant k with absolute temperature T (°K) can be expressed by the following equation [32]:

            k = Ae~~Rr            (1)

where A is the pre-exponential or frequency factor, R is the universal gas constant and E is the activation energy for the reaction. The logarithmic form of equation 1 is:

logk = logA- 1 E (2)

Although the Arrhenius equation applies strictly to chemical reactions, rate constants are often determined in paper ageing studies by following deterioration rates of physical properties such as folding endurance at different temperatures. The fold test is popular since it is very sensitive to accelerated ageing. Plots of the log of the rate constants against inverse temperature (1/T) are then generated. If the plots are linear the Arrhenius equation is said to be valid and activation energies are then determined from the slope of the plot using equation 2. This then allows one to determine the rate constants at any temperature (e.g. room temperature). Following this approach, activation energies in the range of 20-30 kcaUmole have been obtained [22,33-36]. Browning has noted that these values are consistent with activation energies found for the oxidation and hydrolysis of cellulose [22].

Considering the multitude of chemical reactions that may occur during ageing, and the complex relation between the chemical reactions and the physical properties, it is surprising that the Arrhenius plots of folding endurance are still linear. Luner [71 and Erhardt [371 have noted this and have suggested that it may be due to an averageing of the activation energies of several simultaneous chemical reactions, but this is not proven. In addition, different activation energies have been obtained by using different physical or chemical properties to follow the degradation process [38].

Oxidation

A number of workers have noted that degradation of paper during accelerated ageing proceeds faster in air than in nitrogen t62,63], implying that oxidation reactions may be responsible for the deterioration. Recently Arney and Jacobs studied the influence of atmospheric oxygen on the rate of yellowing and on the tensile strength change of a newsprint and rag paper over a range of temperature and relative humidities t64]. They found that yellowing and tensile strength loss varied linearly with oxygen concentration, but did not go to zero in the absence of oxygen. Their suggestion was that two reactions take place, one requiring oxygen, the other independent of oxygen.

One of the difficulties in establishing whether oxidation reactions are a direct cause of strength loss in paper is that oxidative and hydrolytic reactions are cooperative in nature t65]. For example, when the hydroxyl groups of the anhydroglucose unit of cellulose are oxidized, a significant increase in the rate of hydrolysis occurs t66]. Added to this complication is the fact that functional groups introduced through oxidation reactions may also participate in cross­linking reactions.

It is not clear to what extent oxidation reactions occur during natural ageing. Recent work by Arney and Jacobs has shown that the contribution of atmospheric oxidation to the overall rate of deterioration of paper decreases as the temperature is lowered [26]. This implies that under natural ageing conditions the loss of mechanical permanence strictly due to oxidation reactions may be minimal.

Crosslinking

The view that cross-linking reactions may occur during ageing is quite recent. It mainly stems from the work of Back showing that when paper is heated to temperatures up to 350°C, its wet strength increases t671. Back postulates that the increase in wet strength is due to the formation of hemiacetal cross-links. Since this work, increases in wet strength due to accelerated ageing have been attributed to the formation of cross-links t25,56,68,69]. No direct chemical evidence for the formation of such crosslinks have been presented.

It is not certain whether crosslinking reactions occur under conditions of natural ageing, i.e. at ambient temperatures and humidities. The relative importance of hemicellulose and lignin in such reactions is unknown, but recent work on pure cellulose samples show no significant increase in wet tensile strength with ageing [61].is not known whether increased SO2 sorption leads to a loss of mechanical permanence.

Nitrogen dioxide (N°2) present in the atmosphere can also have a negative effect on the mechanical permanence of paper. Smith has speculated that nitrogen dioxide hydrates to yield nitric acid which can degrade cellulose t56]. Recently Daniel and co-workers examined the effect of sulphur and nitrogen dioxides on deacidified paper t57]. They found that compared to control samples certain deacidified papers absorb more acidic pollutants and show a greater loss of pulp viscosity and strength. If these results are confirmed, they have important implications for the use of deacidification treatments for conservation purposes.

Early studies by Herzberg have shown that certain writing inks can have a negative influence on the mechanical permanence of paper t58]. Herzberg has linked the loss of permanence to sulphuric acid formed from the iron sulphate used in the manufacture of the ink. Writing inks are not in common use today; however a variety of printing methods using inks of different chemical composition exist. Of particular importance may be the effect of acidic fountain solutions in the offset printing process. To date, no systematic research has been carried out on the effect of various printing inks or fountain solutions on the mechanical permanence of paper.

There is speculation in the literature that acid groups introduced into the hemicellulose and lignin components of the cell wall during pulping and bleaching processes may lead to a loss of mechanical permanence t59]. This is however a misunderstanding of the physical chemistry of the system. At a neutral pH these groups are completely neutralized by cations such as sodium, calcium and magnesium in the mill white water system. Recent evidence in fact shows that paper made from highly sulphonated lignin-containing pulp such as chemithermomechanical pulp (CTMP) has good mechanical permanence provided the sulphonic acid groups are neutralised t60].

The precise mechanisms by which acid hydrolysis takes place during ageing are not known. Part of the reason for this is a lack of knowledge of the hydrogen ion concentration in the fibre wall during the ageing process. Some researchers draw a parallel between the heterogenous acid hydrolysis of fibres in an aqueous acid solution and hydrolysis occurring during ageing tl0,11]. The hydrolysis is said to take place rapidly in the amorphous regions of the cellulose then slows down in the crystalline regions. However recent work conducted at Paprican shows that the acid catalyzed hydrolysis of pure cellulose during accelerated ageing at 90°C and 80% RH proceeds in a "random" fashion i.e. each chain is cut with equal probability irrespective of its length [61]. Further studies are needed to clarify these effects.

Permanence of Mechanical Pulps

Mechanical pulps are excluded from standards for permanence due to the belief that these pulps are highly unstable. The negative view of the role of mechanical pulps on paper permanence is not new; in 1898 it was commented that paper made from groundwood pulp disintegrates and discolors within a short period of time [78]. During this time groundwood-containing furnishes were sized with alum-rosin which produced an acidic sheet. Therefore it is highly likely that the acidic papermaking conditions were the cause of the loss in mechanical permanence rather than the pulp itself. Unfortunately no one attempted to resolve this issue and mechanical pulps were simply not considered in the early studies on permanence.

Recently Kim et al. have shown that when groundwood pulp is mixed with bleached kraft pulp in increasing amounts the rate of loss in folding endurance with accelerated ageing remains the same [79]. Eggle and coworkers have shown that papers made from bleached kraft pulp have lower mechanical permanence than those made from groundwood [80]. Similar conclusions have also been reached by Oye, who found that degradation rates of MIT folding endurance upon accelerated ageing of newsprint were quite low compared to wood-free papers [81]. In the same study, Oye investigated the effect of lignin content on permanence. Thermomechanical pulp (TMP) was partially delignified with chlorine dioxide to different lignin contents, handsheets were then made at a constant pH of 5.5 and aged. The degradation rates based on the MIT fold test increased with decreasing lignin content.

With the increasing popularity of chemithermomechanical (CTMP) pulps for the production of specialty grades, there has been interest in the permanence of these pulps. Abadie-Maumert and Soteland recently conducted studies on the optical and mechanical permanence of handsheets and commercial papers prepared from unbleached and bleached CTMP, groundwood, bisulphite and a bleached kraft pulp [60]. Their results showed that at a similar pH the rate of loss of tear strength with accelerated ageing was lower for CTMP and groundwood pulps. Though the mechanical pulp­containing papers turned yellow upon ageing they were still perfectly legible and were copied with ease t60]. Johnson and Bird recently showed that when a bleached CTMP is added in increasing amounts (up to 50%) to a bleached hardwood kraft pulp in the presence of calcium carbonate, papers with good mechanical and optical permanence are obtained [82].

THE ROLE OF PULP TYPE ON PAPER

The view from anecdotal evidence is that cotton or rag fibres have good permanence, chemical woodpulp fibres are not as good and mechanical pulp fibres are disastrous. There is little scientific evidence in the literature to support this view.

Perxnanence of Chemical Wood Pulps

In a number of experiments, the ageing characteristics of papers with different fibre composition have been commented on. In early work, Hall showed that bleached softwood sulphite handsheets lost a greater proportion of fold strength than rag handsheets during ageing tl2]. Rasch made similar comparisons for rag pulps, "purified" wood pulps (high a:-cellulose pulps), bleached sulphite pulps and soda pulps, and for a wide range of commercial papers t15]. Rasch noted a strong correlation between <-cellulose content and retention of fold. Other workers showed the same [45,70], leading a Tappi committee to suggest that a:-cellulose content and copper number should be used in specifying pulps for permanent papers tl3].

By this stage, however, it had been noted that manufacturing processes such as alum-rosin sizing and certain pulping and bleaching processes which rendered the sheet acidic influenced permanence, rather than the fibre itself t71]. From then on, less emphasis was placed on fibre "purity" and the furnish mix, and more on the conditions of pulp and paper manufacture t30,72-74]. By 1981, Kelly and Weberg of the U.S. Library of Congress, were able to say, in commenting on specifications and test methods for alkaline papers, that "fully bleached wood pulp fibres appear to be about as stable as cotton or linen in accelerated ageing studies" t75].

Today one can review this trend, from its initial insistence on high a:­cellulose content to the acceptance of any bleached wood pulp, and one can relate it to the relationship between fibre strength and chain length t76]. Cotton fibres have high ~:-cellulose content but also a high degree of polymerisation (DP). Residual acidity would take some time to reduce the DP of cotton to a point where serious loss of fibre strength was evident. Bleached wood pulp, of lower DP than cotton, has initially satisfactory fibre strength, but residual acidity would reduce the DP to a dangerously low level in a shorter time than cotton. Above pH 7, with no acid attack, the DP of wood pulp should remain sufficiently high to give good fibre strength and hence good retention of mechanical properties. Indeed, the initial DP of the pulp was recognised as a very important aspect of permanence by Faulhaber and Pietrzyk almost forty years ago t77]. They showed that an unbleached and bleached kraft pulp of high initial DP (DP=1000-1200) were as stable as paper made from rag stock.

The general recommendation that as little rosin as possible should be used, can be explained partly by the desire to avoid discolouration, and partly by the fact that any excess of rosin, above the amount that will react with the added alum would cause undesirable production problems such as foam and press roll sticking. It was therefore traditional to add an excess of alum to avoid these issues. In light of this, the results of Shaw and O'Leary can be explained. They found that adding alum to stock containing rosin caused less deterioration of folding endurance and tear strength than when rosin was absent [20]. They termed this a "protective effect" of rosin. Presumably, in the former case, rosin was able to react with alum to form aluminium rosinate, thus reducing the amount of unreacted alum in the dry sheet.

Over the years, more efficient forms of rosin size have been developed which require less alum and therefore operate in the range 6-6.6 pH. While this can only help ageing characteristics, these sizes still use alum as a fixing agent and leave the paper acid. Alternative systems have recently been introduced which enable rosin to be used at neutral t88,89] or alkaline pH [90] without any alum.

Considering that rosin sizing is still widely used after 140 years and that in commercial operations aluminum compounds are invariably involved, it is a matter for some astonishment that we still do not know for certain, either a) the exact mechanism by which aluminum compounds affect permanence, or b) whether rosin itself contributes to loss of permanence.

Synthetic Sizing

Calcium carbonate fillers cannot be used to any great extent in acid systems. Thus, in the 1940's and '50s pressure arose to develop a sizing system which could operate in the neutral or alkaline regime. The Hercules Powder Company introduced "Aquapel" based on an alkyl ketene dimer (AKD) [91,92] and since then other sizing systems have been invented, such as one based on an alkenyl succinic anhydride (ASA) t93]. The chemistry of these systems has been covered well by Roberts [94].

With regard to ageing characteristics, while it is generally assumed that an AKD size is beneficial, the evidence is entirely based on work with AKD/calcium carbonate systems. The influence of AKD itself has not been investizated, nor has that of ASA size.

Nowadays, 0.25-0.75% alum is often added to alkaline systems to improve sizing, drainage and machine runnability [95]. This is a similar amount to that used in rosin/alum sizing. The significance to permanence of this amount, with or without calcium carbonate loading has not been investigated.

Permanence of Recycled Pulps

With increased production of recycled pulps there is growing intsest in their permanence. There are only two references in the literature which address the permanence of recycled papers [83,84]. In the first study it is found that a 100% deinked book paper stock shows a 93% retention of folding endurance after 72 hours of ageing at 105°C indicating an unusually good stability t83]. The authors attribute this to the "purifying" effect of the mild alkaline treatment during the deinking process. More recently, Ernst has studied the ageing properties of fine paper grades including recycled papers t84]. He concludes that recycled papers do not age more rapidly than papers made from virgin stock.

G ADDITIVES ON PERMANENCE

THE EBECT OF PAP]

Papermaking involves the use of several non-fibrous materials. While some of these additives have a proven connection with ageing performance, many others have never been investigated even though some link to permanence could be conjectured.

Rosin Sizing

Rosin sizing superceded earlier sizing treatments in which the sheet was impregnated with animal glue such as gelatin. Rosin size was prepared as the sodium salt of abietic acid, the principal component of wood resin. This was precipitated by alum onto the fibres in the wet end of the papermachine. Typically, 2-3% of rosin (on weight of fibre) was used, which required approximately 4% alum, giving a stock pH of 4-5 and considerable residual acidity in the sheet.

This acidity was quickly recognized as a potential factor in paper ageing. Hoffman for example reported work by Hall which recommended that the alum addition should not exceed 80% of the minimum necessary rosin addition t85]. Hoffman's own work also emphasized the importance of residual acidity; a minimum extract pH of 4.8 was suggested, but, like Hall, Hoffman made no recommendation concerning the level of rosin addition t85]. Rasch et al. went further; while they came to similar conclusions regarding alum use and pH, they further suggested that rosin itself had no deleterious effects on permanence [86]. Other workers have also concentrated on the alum component of rosin/alum sizing [17,87]. Rosin, if criticized at all, was criticized for its notential to discolour during ageing t21].

Surface Sizing

Before the advent of internal sizing, printing and writing papers were sized using animal glues such as gelatin to reduce water absorbency, to provide a sheet surface which would allow good definition of printing or writing, and to improve erasability. At an application rate of between 3 and 7% by weight on fibre, this material had the added advantage of strengthening the sheet. This led to the general view that gelatin sized papers have good permanence, although concern was expressed that alum, which came to be used to control size viscosity, to act as a size preservative and to harden the sized sheet, could negate some of the benefits of gelatin size [19,105]. Gelatin sizes are rarely used today. However, starch is frequently applied on machine at the size press in conjunction with internal size, at an addition rate of between 2 and 5%. Early studies have shown that starch addition has no adverse affect on permanence [16,44].

Coating

Literature on the permanence of coated paper is almost non-existent. The archival qualities of coated papers were examined by Shahani and McComb [106]. They found that after coating an alkaline sized base paper, the rate of decrease of fold endurance increased, even though the rates for tear and tensile strength were unaffected. This was true even for a coating which was alkaline. They were unable to explain their data. Recently, the committee revising the ANSI standard commissioned experimental work and found that an slkaline coating was not sufficient and that the base paper itself had to be acid-free rlo7l.

THERELAITVEUJPORTANCEOFANECDOTAL,HETORICALAND SC~BUIRCEVIDENCE

It is clear that the quality of evidence in the literature on ageing is highly variable. At one end we have the purely anecdotal evidence. These are observations which associate permanence, with certain other qualities as a result of an isolated case or general experience. An example might be the widespread belief that if paper yellows with age it also necessarily becomes embrittled. A more sophisticated form of this approach is the careful scientific examination of historical documents. This is best exemplified by the work of Hansen in 1939 [108]. He found that a book published in 1576 was made from different papers, some of which had aged badly, while others were well preserved. He found that the well preserved sheets contained at least 2% calcium carbonate and had a pH of at least 7.5. He concluded that this was the recipe for permanent paper.

Fillers

Non-Alkaline Fillers

It is not immediately apparent how non-alkaline fillers could help permanence, and perhaps for that reason they have rarely been considered in the literature. Howeva, in a systematic study of accelerated ageing on 72 experimental papers manufactured by the National Bureau of Standards, clay, titanium dioxide (TiO2) and zinc sulphide were considered, no effects on mechanical permanence were observed [20]. Kelly and Weberg, however, commented on the possible adverse effect of TiO2 [75]. They cited the work of Edgerton [96] who published studies showing that in W light and at high humidity, TiO2 catalysed the oxidation of textiles. Kelly and Weberg say that the effect has been shown to apply to paper also though no data were given [77,97]. No other work on non-alkaline fillers has been found.

Alkaline Fillers

The early identification of acidity as a cause of impermanence led to the suggestion that an alkaline buffer in the form of calcium carbonate would be beneficial to neutralize any acid that may be generated in the ageing process. Subsequently, the work of Shaw and O'Leary confirmed the benefits to permanence of using calcium carbonate [20]. In their work the effect of a precipitated calcium carbonate and 4 ground calcium carbonates were examined. All markedly improved the permanence characteristics, and this was the case whether or not rosin size and alum were also used. Shearon confirmed these findings in 1957 [98]. Considerable work was done at the Barrow laboratory in the 1950's and 1960's in establishing the value of calcium carbonate for improved mechanical permanence [99]. This led to the concept of an "alkaline reserve" of either calcium or magnesium carbonate in the sheet to prevent loss of strength properties with ageing. Since then other workers have demonstrated the value of using calcium carbonate for increased permanence [81,100-103] .

Calcium carbonate is available in two forms - as ground calcium carbonate produced by grinding a mined form of calcium carbonate such as marble, limestone or chalk, and as precipitated calcium carbonate (PCC) produced by carbonation of lime. The differences of importance to the papermaker are principally ones of particle size, shape and abrasivity, and of cost. Natural calcium carbonate is brighter than clay as well as being cheaper in some parts of Europe, and this has been a factor in promoting its use in European fine paper mills. Precipitated calcium carbonate as produced on site, has a pH of approximately 8.5 but it is reported to contain unreacted lime inside the carbonate particles, and this may account for reports of the sheet pH rising to between 9 and 10 tl04]. A potential danger, therefore, is of alkaline hydrolysis of the fibres; this has not been examined.

ACKNOWLEDGEMENTS

The authors wish to thank the National Archives of Canada for funding a post-doctoral fellowship for one of the authors (X. Z.), Dr.K. Hendriks of National Archives of Canada for stimulating discussions and Dr.R. W. Johnson of Du Pont Chemicals for providing key literature references.

REFfERACES

[1] Communications Canada, Information Services, News Release: "Government to Print Publications on Permanent Paper", January 15.1992.

[2] American National Standards Institute, American National Standard for Information Sciences - Permanence of Paper for Printed Library Materials. ANSI Z39.48-1984.

[3] American National Standards Institute, American National Standard for Information Sciences - Permanence of Paper for Printed Library Materials. ANSI Z39.48-1992.

[4] Draft international Standard (DIS) 9706 in preparation, "Papers for Long­Life Documents, Records and Publications-- Specifications for Permanence.

[S] ASTM Standard D3290-86, "Standard Specification for Bond and Ledger Papers for Permanent Records".

[6] Neimo, L., "Accelerated Heat Aging of Cellulose", Paperi ja Puu 46(1):7 (1964).

[7] Luner, P., "Paper Permanence", Tappi 52(5):796 (1969).

[8] Roberson, D. D., "The Evaluation of Paper Permanence and Durability", Tappi 59(12):63 (1976).

[9] Wilson,W. K., and Parks,E. J., "Historical Survey of Research at the National Bureau of Standards on Materials for Archival Records", Restaurator 5: 191 (1983).

[10] Feller,R. L., Lee, S. B., and Curran, M., "Three Fundamental Aspects of Cellulose Deterioration", Art and Archaeology Technical Abstracts, 22 (1):278 (1985).

249

In the absence of other data these approaches are the only basis for the development of ageing standards. However, from a scientific viewpoint they are quite inadequate. First they are uncontrolled. The conditions under which different books or papers have been kept is unknown. Furthermore in such studies the experimenter is at the mercy of historical correlations. For example the development of stone groundwood as a furnish followed the wide acceptance of alum-rosin sizing so that virtually all sheets made from stone groundwood from 1850 to 1950 contain alum. It is thus not clear from historical data whether the poor permanence of groundwood-containing sheets is caused by the furnish or the alum. Finally the historical literature is silent on the question of new furnishes. It cannot tell us whether a fully bleached kraft sheet will last 100 years since the bleaching of kraft pulp was not widely carried out until the 1940's. It cannot tell us whether a TMP sheet will last 50 years since TMP was not manufactured until the 1960's. By contrast a well designed scientific experiment controls the variables of manufacture and ageing systematically so that upon completion conclusions can be made that are unequivocal. In addition it allows for the development of theories for the mechanism of the property changes that occur upon ageing, so that the work gains scientific credibility.

This review of the evidence contains an inherent bias. It is the belief that those results obtained by systematic scientific experiments carry a greater weight than should be afforded to anecdotal evidence. While note may be taken of anecdotal evidence, and attempts made to explain it, such evidence should not control the basis upon which decisions are made. The conclusions drawn below have been made by placing more emphasis on the evidence that has the highest credibility.

CONCLUSIONS

A review of the existing literature on paper permanence points towards acid hydrolysis of the cellulose as the primary cause of paper embrittlement. The sources of acidity in the sheet have been linked mainly to the use of alum during the rosin sizing process and to atmospheric contaminants. It has been shown that if an "alkaline reserve" is present in the form of calcium carbonate a paper ages well.

This literature review produced no scientific evidence to explain why mechanical pulps and unbleached chemical pulps are excluded from the standards for mechanical permanence. This seems to arise only from anecdotal evidence. There is substantial scientific evidence to indicate that as long as the sheet manufacturing conditions are neutral to alkaline with an alkaline reserve of calcium carbonate, mechanical pulps have excellent retention of strength properties upon ageing.

[24] Parks, E. J. and Hebert, R. L., wAccelerated Aging of Laboratory Handsheets: Reflectance, Moisture Regain, Sonic Modulus and Differential Thermal Analysis", NBS Report 10-687, Feb.22,1972, NTIS COM 75 10602.

[25] Graminski, E. L., Parks, E. J., and Toth, E. E., ~The Effects of Temperature and Moisture on the Accelerated Aging of Paper" in "Durability of Macromol. Materials", Ed. R. K. Eby, ACS Symp.Sa. 95, Chap.24, (1979).

[26] Arney, J. S., and Jacobs, A. J., "Newsprint Deterioration: The Influence of Temperature on the Relative Contribution of Oxygen-Independent and Oxygen-Dependent Processes in the Total Rate", Tappi, 63 (1):75 (1980).

[27] Tappi Provisional Method T453 pm-85, "Effect of Dry Heat on Properties of Paper", 1985.

[28] Tappi Provisional Method T544 pm-85, "Effect of Moist Heat on Properties of Paper and Board", 1985.

[29] ISO Standard 5630/3-1986 (E), "Paper and Board- Accelerated aging­Part 3: Moist heat treatment at 80°C and 65So relative humidity".

t30] Wilson, W. K., Harvey, J. L., Mandel, J., and Worksman, T., "Accelerated Aging of Record Papers Compared with Normal Aging", Tappi 38(9):543 (1955).

[31] Wilson, W. K., and Parks, E. J., "Comparison of Accelerated Aging of Book Papers in 1937 with 36 Years of Natural Aging", Restaurator 4:1 (1980).

[32] Arrhenius, S. Z., Phys.Chem. 4:226 (1889).

[33] Stamm, A. J.,"Thermal Degradation of Wood and Cellulose", Ind.Eng.Chem. 48(3):413 (1956).

[34] van Royen, A. H., "Le vieillisement artificiel a 100°C compare au vieillisement naturel a temperature ambiante", Assoc.Tech.Ind.Papetiere, Bull. C No.6:223 (1957).

[35] Millet, A. M., Western, L. J., and Booth, J. J., "Accelerated Aging of Cellulosic Materials: Design and Application of a Heating Chamber", Tappi 50(11):74A (1967).

t36] Gray, G. G., "An Accelerated-Aging Study Comparing Kinetic Rates vs. TAPPI Standard 453", Tappi 52(2):325 (1969).

[11] Fellers, C., Ivasen, T., Lindstrom, T., Nilsson, T., and Rigdahl, M., in "Aging/Degradation of Paper- A Literature Survey", FOU-projektet for papperskonservering, Report No. 1E, ISSN 0284-5636, Stockholm, Sentember 1989.

[12] Hall, G., "Permanence of Paper", Technical Assoc. Papers, Series IX, p.92, June,1926.

[13] Chrk, F. C., "Permanence and Durability Committee Report", Tech. Assoc. Pasers 18:470 aune, 1935).

[14] Herzberg, W., "Influence of High Temperatures on the Strength of Paper", Papier-Ztg. 24, no.61:2343 (1899).

[15] Rasch, R. H., "Quality of Purified Wood Fibers", Paper Trade J., 88, no.8:233 (Feb.21,1929).

[16] Rasch, R. H., Shaw, M. B., and Bicking, G. W., "Highly Purified Wood Fibas as Paper-Making Material", Bureau of Standards J. Res. 7, no.5:765 (Nov.1931).

[17] Jarrel, T. J., Hankins, J. M., and Veitch, F. p., tThe Effect of Inorganic Acids on the Physical Properties of Waterleaf Rag Bond Paper", Paper Trade J. 95, no.l4:28 (Oct.6,1932).

[18] Rasch, R. H., and Scribner, B. W., "Comparison of Natural Aging of Paper with Accelerated Aging by Heating", Bureau of Standards J.Res. 11, no.6:727 (Dec.1933).

[19] Richter, G. A., "Accelerated Aging Tests for Determining Pamanence of Papers", Ind.Eng.Chem., 26(11):1154 (1934).

[20] Shaw, M. B., and O'Leary, M. J., "Effect of Filling and Sizing Materials on Stability of Book Papers", J.Res.National Bureau of Standards 21, no.5:671 (Nov.1938).

[21] Scribner, B. W., "Degradation of Sulphate Fiber by Bleaching" Paper Ind. 19:1279 (Feb.1938).

[22] Browning, B. L., and Wink,W. A., "Studies on the Permanence and Durability of Paper I. Prediction of Permanence", Tappi 51(4):156 (1968).

[23] Cardwell, R. D., "Aging of Paper", Ph.D. Thesis, N.Y. State College of Forestry, Syracuse, N.Y., 1973.

[50] Langwell, W. H., "The Permanence of Paper- Part IV", Proc.Tech.Sect. Brit.Paper and Board Makers Assoc. 36(1):199 (1955).

[51] Langwell, W. H., "The Permanence of Paper- Part vn, Ibid., 37(3):495 (1956).

[52] Langwell, W. H., "The Permanence of Paper- Part VI", Ibid., 40(1):105 (1959).

[53] Hudson, F. L., and Milner, W. D., "The Permanence of Paper- The Use of Radioactive Sulphur to Study the Pick-Up of Sulphur Dioxide by

Paper", Paper Technol. 2, no.2:155 (1961).

[54] Edwards, C. J., Hudson, F. L., and Hockey, J. A., "Sorption of Sulphur Dioxide by Paper", J .Appl. Chem. 18(5) : 146 (1968) .

[55] Hudson, F. L., Grant, R. L., and Hockey, J. A., "The Pick-Up of Sulphur Dioxide by Paper", J.Appl.Chem. 14(10):444 (1964).

[56] Smith, R. D., "Paper Impermanence as a Consequence of pH and Storage Conditions", Lib.Quarterly 39, no.2:153 (April 1969).

[57] Daniel, F., Flieder, F., and Leclerc, F., "The Effects of Pollution on Deacidified Paper", Restaurator 11:179 (1990).

[58] Herzberg, W., "Destruction of Paper by Writing Ink", Paper Trade J. 76, no.l0:55 (March 8,1923).

[59] Luner, P., "Evaluation of Paper Permanence", Wood Sci.Technol. 22:81 (1988).

[60] Abadie-Maumert, F. A., and Soteland, N. A., "Resistance des CTMP au Vieillissement", Revue A.T.I.P. 44(6):223 (1990).

[61] Zou, X., Gurnagul, N., Uesaka, T., and Bouchard, J., "Accelerated Aging of Papers of Pure Cellulose: Mechanism of Embrittlement", to be published in the Journal of Pulp and Paper Science.

[62] Lewis, H. F., "The Relative Stabilities of Rag and Purified Sulphite Pulps", Paper Trade J. 95, no.21:29 (Nov.24,1932).

[63] Parks, E. J., and Hebert, R. L., "Accelerated Aging of Laboratory Handsheets: Changes in Acidity, Fiber Strength, and Wet Strength", National Bureau of Stand. Report 10627, Dec.27,1971.

[371 Erhardt, D., "Paper Degradation: a Comparison of Industrial and Archival Concerns", in the proceedings of the Tappi Paper Preservation Symposium, Washington, D.C. Oct.19-21,1988, Ed.P.Luner, p.63.

[38] Saad, S. M., and El-Kholy, A. E., "Kinetic Investigation of Thermal Aging of Unbleached Egyptian Kraft Bagasse Paper-Pulp", Holzforschung, 34 (4):147 (1980).

[39] Wilson, W. K., and Parks, E. J., HAn Analysis of the Aging of Paper: Possible Reactions and their effects on Measurable Properties", Restaurator 3:37 (1979).

[40] Barrow, W. J.,and Sproull, R. C., "Permanence in Book Papers", Science 129,no.3356:1075 (April 24,1959).

[41] Edlund, T.,"Schwefelsaure Im Papier", Papier Fabrik. 8:765 (1910). [original not seen].

[42] Hoffman, W. F., "Deterioration of Paper by Excessive Acidity", Paper Trade J. 88 no.12:59 (1929) .

[43] Parks, E. J.,"Thermal Analysis of Modified Cellulose", Tappi 54(4):537 (1971).

t44] Richter, G. A.,"Durability of Purified Wood Fibers: Accelerated Aging Tests of Various Types of Paper-Making Fibers", Ind.Eng.Chem. 23(4):371 (1931).

[45] Kimberly, A. E.,"Deteriorative Effect of Sulphur Dioxide upon Paper in an Atmosphere of Constant Humidity and Temperature", Bureau Stand.J.Res. 8(2):159 (1932).

[46] Jarrel, T. D., Hnnkins, J. M., and Veitch, F. P., HDeterioration of Paper as Indicated by Gas Chamber Tests", U.S. Dept.Agriculture Tech.Bull. no.605, January,1938.

[47] Langwell, W. H.,"The Permanence of Paper - Part I", Tech.Bull.Brit. Paper and Board Makers' Assoc. 29(1):21 (1952).

[48] Langwell, W. H., "The Permanence of Paper- Part II", Ibid., 29(2):52 (1952).

[49] Langwell, W. H., HThe Permanence of Paper- Part III", Ibid., 30(6):170 (1953).

[771 Faulhaber, M., and Pietrzyk, K., "Uber die Dauerhaftigkeit von Papier", Wochbl. Papierfabrik., 84, no.5:147 (March 1956).

[78] Society of Arts.- Committee on the deterioration of paper. (Report of the committee on the deterioration of paper) in: J.Soc.Arts (London) 46:597 (1898).

[79] Kim, B. Y., Isogai, A., Onabe, F., and Usuda, M., "Photochemical Reactions of Paper", in "Wood Processing and Utilization", J. F. Kennedy, G. O.Phillips, and P. A.Williams ed., John Wiley and Sons, N.Y. 1989, chapter 50, p.381.

[80] Eggle, R., Hofer, H. H., and Scherer, B. K., "Effect of Stock Composition, Type of Filler and pH Value on the Aging Resistance of Paners", Wochenbl.Papierfabrik., 112, no.22:809 (November 1984).

[81] Oye, R., Okayama, T., Matsubara, H., and Hanamura, K., "Studies on Degradation of Paper and Conservation II- Degradation of Mechanical Pulp Containing Paper", Japan Tappi 45(2):277 (1991).

[82] Johnson, R. W., and Bird, A., "CTMP in Fine Papers: Impact of CTMP on Permanence of Alkaline Papers", Preprints of Tappi Papermakers Conference, Seattle WA, April 8-10, 1991, p.267.

[83] Anonymous, "Stability of Paper made from De-inked Paper Stock", News.Bull., Paper Section, National Bureau of Standards: 4-5 (April, 1941).

[84] Ernst, U.,"Investigations on the Aging Properties of Writing, Printing, and Copying Papers", Das Papier 43(11):616 (1989).

[85] Hoffman, W. F., "The Effect of Residual Acid on Rate of Deterioration of Paper", Paper Trade J., 86,no.9: 58 (March 1,1928).

[86] Rasch, R. H., Shaw, M. B. and Bicking, G. W.,"A Study of Some Factors Influencing the Strength and Stability of Experimental Papers Made from Two Different Sulphite Pulps", Bureau of Standards J.Res., 11. no.l:7 (July 1933).

[87] Strachan, J.,"Chemical Residues in Paper", Papermaker and British Paper Trade J., 89, no.5: TS65 (May, 1935).

[88] Herner, B., "Neutral Sizing System Allows Fast, Easy Conversion for Acid Machines", Pulp Paper 64(1):141 (1990).

[64] Arney, J. S., and Jacobs, A. J., "Accelerated Aging of Paper: The Relative Importance of Atmospheric Oxidation", Tappi 62(7):89 (1979).

[65] Lindstrom, T., Discussion Contribution: "Slow Fires- It's Paper Chemistry, Physics and Biology", in the proceedings of the Tappi Paper Preservation Symposium, Washington, D.C. Oct.19-21,1988, Ed.P.Luner, p.74.

[66] Daruwalla, E. H., and Narsian, M. G., "Detection and Identification of Acid-Sensitive Linkages in Cellulose Fiber Substances", Tappi 49(3):106 (1966).

[67] Back, E. L., "Thermal Auto-Crosslinking in Cellulose Material", Pulp Paper Mag.Can. 68(4):T165 (1967).

[68] Graminski, E. L., "The Stress-Strain Behaviour of Accelerated and Naturally Aged Papers", Tappi 53(3):406 (1970).

[69] Koura, A. and Krause, T. H., HAging of Paper I- Studies of the Aging of Lignin-free Pulps", Papier 31, no.lOA: V9 (Nov.1977).

[70] Burton, J. O., "Permanence Studies of Current Commercial Book Papers", Bureau Standards J. Res., no.3:429 (Sept.1931).

[71] MacNaughton, W. G., "Fiber Composition and Paper Durability", Pulp Paper Mag.Can. 25, no.l5:469 (April 14,1927).

[72] Arnould, E., "La Fabrication des Papiers de Longue Conservation et Causes de leur Alteration". Le Papier 31(4):431 (1928).

[73] Klemm, P., "Permanence and Durability of Paper", Wochbl. Papierfabr. 63, no.32:607-609; no.33:625-627 (Aug.6,13, 1932).

[74] Sclawy, A. C., and Williams, J. C., "The Demand for ALlcaline, Permanent, Paper", Preprints of Tappi Papermakers Conf. April S 8,1981, Chicago, IL, page 85.

[75] Kelly, G. B., and Weberg, N., HSpecifications and Methods of Test for Alkaline Papers", Tappi Papermakers Conf. April 6-8,1981, Chicago, IL, page 71.

[76] Gurnagul, N., Page, D. H., and Paice, M. G., "The Effect of Cellulose Degradation on the Strength of Wood Pulp Fibres", Nordic Pulp and Paper Res. J., (3):152 (1992).

by Changes in its Chemical Properties", Vop. Dolgovechnosti Dokumenta 1973: 16-18.

[103] Dessauer, G., "Aging of Paper and Modern Methods of Prevention", Wochbl. Papierfabr. 103 (23-24):902-906 (1975).

[104] McCrady, E.,"The Trouble with ALkaline Chemistry", Alk.Pap. Advocate 5, no.2:12 (May 1992).

[105] Baker, W. E. B., "Relationship of Impurities to Paper Permanence", Paper Trade J. 91, no.4:51 (July 24, 1930).

[106] Shahani, C. J., and McComb, R. E., "A Clarification on Specifications for Archival Paper", Tappi 70(9):128 (1987).

[107] Humprehys, B., and Kalina, C. M., "Revising the American National Standard for Permanence of Paper (ANSI Z39.48-1984): Changing Market Factors, Changing Paper Technology, and New Research Questions", Tappi Papermakers Conference, Seattle WA, April 8­10.1991, p.243.104.

[108] Hanson, F. S., "Resistance of Paper to Natural Aging", Paper Ind. and the Paper World 20: 1157 (February 1939).

 [89] Roberts, J. C., "Neutral Sizing", published by PIRA, Leatherhead, UK, (1992).

[90] Biermann, C. J., "Rosin Sizing with Polyamine Mordants from pH 3 to 10", Tappi 75(5):166 (1992).

[91] Weisgerber, C. A., Davis, J. W., and Roberson, W. H., "A New Sizing Agent for Paper- Alkylketene Dimers", Tappi 39(12):21 (1956).

[92] Espy, H. H., "The Genesis of Alkaline Sizing and Alkaline-curing Wet­Strength Resins", Alk.Pap. Advocate 3, no.3:28 (Aug.1990).

[93] Wurzberg, O. B., and Mazzarella, E. D., U.S. pat. 3,102,064 (Aug.27,1963).

[94] Roberts, J. C., "Paper Chemistry", published by Chapman & Hall, New York, 1991.

[95] Wortley, B., "Alum use in ALcaline and Neutral Sizing Systems", PIMA Maz. March 1988.

[96] Edgerton, G. S., "The Role of Hydrogen Peroxide in the Photochemical Degradation of Cotton Sensitized by Vat Dyes and some Metallic Oxides", J.Text.Inst. 39:T 305, (1948).

[97] Kelly, G. B., and Williams, J. C., "Inhibition of Light Sensitivity of Papers Treated with Diethyl Zinc", paper presented at the 178th ACS meeting, Washington D.C., Sept.10-12,1979.

[98] Shearon, W. H., Jr. "The Old Gray Book", Ind.Eng.Chem. 49, no.ll: 25-26A (Nov.1957).

[99] Virginia State Library, "The Manufacture and Testing of Durable Book Papers Based on the Investigations of W.J.Barrow". Edited by R. W.Church, Richmond VA, 1960.

[100] Belaya, I. K., "Aging of Cotton and Flax Fiber Papers under the Simultane Action of Heat and Moisture't, Starenie Bumagi, Akad. Nauk SSSR, Lab. Konserv. Restavratsii Dok. 1965: 4546.

[101] Belaya, I. K., "Changes in some Chemical Properties of Flax and Cotton Papers during Accelerated Aging", Prichiny Razrusheniya Pamyatnikov Pis'mennosti Perchati. Akad. Nauk SSSR, Lab Konserv. Restavratskii Dok. 1967:48-50.