NC State
BioResources
Čabalová, I., Kačík, F., Gojný, J., Češek, B., Milichovský, M., Mikala, O., Tribulová, T., and Ďurkovič, J. (2017). "Changes in the chemical and physical properties of paper documents due to natural ageing," BioRes. 12(2), 2618-2634.

Abstract

The aim of this study was to assess the changes in chemical and physical properties of selected paper documents, dating from the 18th century to the present date. Paper documents from the 18th, 19th, and 21st centuries had a pH ranging from slightly acidic to alkaline values, whereas paper documents from the 20th century were more acidic. The contents of mannose, glucose, holocellulose, and lignin were significantly correlated with the pH of the paper. High contents of both carboxylic acids (acetic and formic) and the most durable monosaccharide (glucose) were found in the oldest examined paper documents originating from the 18th and 19th centuries. The lowest percentage of brightness was found in paper documents originating from 1920 and 1923. The lowest breaking length was associated with papers originating from 1920.

Download PDF

Full Article

Changes in the Chemical and Physical Properties of Paper Documents due to Natural Ageing

Iveta Čabalová,a,* František Kačík,a Ján Gojný,b Břetislav Češek,b Miloslav Milichovský,b Ondřej Mikala,b Tereza Tribulová,a and Jaroslav Ďurkovič c

The aim of this study was to assess the changes in chemical and physical properties of selected paper documents, dating from the 18th century to the present date. Paper documents from the 18th, 19th, and 21st centuries had a pH ranging from slightly acidic to alkaline values, whereas paper documents from the 20th century were more acidic. The contents of mannose, glucose, holocellulose, and lignin were significantly correlated with the pH of the paper. High contents of both carboxylic acids (acetic and formic) and the most durable monosaccharide (glucose) were found in the oldest examined paper documents originating from the 18th and 19th centuries. The lowest percentage of brightness was found in paper documents originating from 1920 and 1923. The lowest breaking length was associated with papers originating from 1920.

Keywords: Paper documents; Natural ageing; pH; Formic acid; Acetic acid; Fibre length; Holocellulose

Contact information: a: Department of Chemistry and Chemical Technologies, Technical University in Zvolen, T.G. Masaryka 24, 960 53 Zvolen, Slovak Republic; b: Institute of Chemistry and Technology of Macromolecular Materials, University of Pardubice, Studentská 9, 532 10 Pardubice, Czech Republic; c: Department of Phytology, Technical University in Zvolen, T.G. Masaryka 24, 960 53 Zvolen, Slovak Republic; *Corresponding author: cabalova@tuzvo.sk

INTRODUCTION

The deterioration of paper is caused by several factors such as acid hydrolysis, oxidative agents, light, air pollution, or the presence of microorganisms (Zou et al. 1996a,b; Zotti et al. 2008; Zervos 2010; Area and Cheradame 2011; Kraková et al. 2012). During natural ageing, loss of paper strength is a result of the degradation of cellulose, its main structural component. Cellulose degradation in paper reaches various modes depending on a variety of chemical processes (acid and enzymatic hydrolysis, alkaline and oxidative degradation), thermal impact (influence of different temperatures), and radiation (ultraviolet and high energy), considering the acid hydrolysis to be most prevailing (Whitmore and Bogaard 1994).

Lignin content is of major interest in the preservation of paper-based materials. It is a complex and partly unstable natural organic polymer, and its limited photostability is important for exhibited objects. The influence of lignin on the stability of cellulose in paper is not well understood, but it may provide a limited antioxidant effect. However, its oxidation may also contribute to an accumulation of acids, which destabilize the material (Zou et al. 1993; Schmidt et al. 1995; Begin et al. 1998; Trafela et al. 2007). A major source of acidity development within paper documents during aging is known to be the hydrolysis of acetate ester groups in the hemicellulose (Polovka et al. 2006; Zervos 2007; Jablonský et al. 2012a).

Paper before ca. 1850 was produced with the traditional sizing process using gelatine (Dupont 2003), which resulted in an approximately neutral or slightly acidic material. Generally, gelatin within old books remains in good condition if it has been kept in a favourable environment in the absence of corrosive inks or biodeteriogens. Gelatine is beneficial to paper and increases its longevity (Strlič et al. 2004; Lichtblau et al. 2008). For paper produced between 1850 and 1990, the pH values of aqueous extracts are frequently lower than 6 due to the application of the alum-rosin size system. This decreases the lifetime expectancy of paper. From 1990 to the present, the production of paper has changed to alkaline sizing, and contemporary paper is once again more stable (Trafela et al. 2007). The reason was to reduce costs. The modern “PCC” calcium carbonate filler, which is required for alkaline pH during paper forming, made it less expensive to make bright paper (Hubbe 2005). The strength of paper is determined by the strength of the individual fibres and the strength of the bonding of these fibres into a network (Page and Seth 1979; Ververis et al. 2004).

According to Paavilainen (1993a,b), there is a decrease in tensile strength corresponding to an increase in fiber coarseness, suggesting that the most important factors for high tensile strength of paper are a good bonding ability and a high intrinsic fibre strength. Tensile strength is determined by the bonded area, both external and internal fibrillations, the amount of fines, and especially by wet fibre flexibility. Gandini and Pasquini (2012) noted that the mechanical properties of paper are directly linked to the capacity of its fibres to establish inter- and intra-fibre bonds. Maximizing the bonded area and fibre length can increase the activation process, while deformations such as kink and curl can decrease activation.

Deformed fibres also lead to a non-uniform load distribution, thereby decreasing the networks ability to carry load because they cause local stress points (Vainio 2007). Mohlin et al. (1996) studied the impact of fibre deformation on sheet strength. The authors reported a decrease in tensile strength and tensile stiffness with an increase in the number of fibre deformations and a decrease in the shape factor.

Analysis of fibre traits such as fibre length, fibre diameter, lumen width, cell wall thickness, and their derived morphological characteristics has become important in estimating the pulp quality of fibre (Dinwoodie 1965; Amidon 1981). Fibre length and strength are particularly important for tearing resistance (Tamolang et al. 1968; Wangaard and Williams 1970; Seth and Page 1988). In addition, Seth (1990) showed that tensile strength is affected by fibre length.

The objectives of this study were to assess the changes in pH values, contents of carboxylic acids, holocellulose and lignin, fibre dimensions, breaking length, and the brightness of paper documents at various ages. The study also serves to clarify significant relationships between the examined traits.

EXPERIMENTAL

Materials

Paper samples were taken from a book printed in each of the years 1719, 1762, 1784, 1807, 1839, 1853, 1859, 1920, 1923, 1956, and 2007. All the researched books were stored in one single place, in an old church in the central part of Slovakia, which means these were undergoing a natural aging process. The publication year was considered to be the production year of the paper document.

Methods

Determination of the pH values

Determination of the pH values of paper surfaces (pHS) was performed according to the Slovak Technical Standard (STN) 500374 (1996) (with a drop of distilled water and subsequently by placing the pH electrode on the paper surface, with sampling at five locations), whereas determination of the pH values of cold aqueous paper extracts (pHE) was conducted according to the standard STN ISO 6588 (50 0381) (1993) (cold extraction, 2 g of paper sample per 100 mL of water, 1 h). The precision in pH determinations was estimated to be less than ± 0.2 units. Measurements were performed on five replicates per sample.

Contents of carboxylic acids

Approximately 2 g of paper were accurately weighed and added to 15 mL of pure water. The suspension was mixed for 2 h and subsequently filtered through a 0.45 μm filter (Jablonsky et al. 2012b). The carboxylic acid (formic, acetic) contents were analysed using high-performance liquid chromatography (HPLC) and a method slightly modified from Sluiter et al. (2008a), as follows: chromatograph, HPLC 1200 (Agilent, Santa Clara, CA, USA); column, Polymer IEX H form (Watrex, Praha, Czech Republic); mobile phase, 9 mM H2SO4; flow rate, 0.5 mL min–1; detector, RI (refractive index); injected volume, 100 μl; and temperature, 35 °C. Results are presented as formic and acetic acids for both free acids and their water soluble salts. Measurements were performed on four replicates per sample.

Contents of saccharides and lignin

Paper samples were hydrolysed using 72% (w/w) sulfuric acid, and both lignin and neutral saccharides were determined according to the NREL method (Sluiter et al. 2010). Quantitative analyses of saccharides were carried out by HPLC under the following conditions: chromatograph, HPLC 1200 (Agilent, Santa Clara, CA, USA); column, Aminex HPX – 87P (Bio-Rad, Hercules, CA, USA); mobile phase, deionized water; flow rate, 0.6 mL min–1; detector, RI; injected volume, 50 μL; and temperature, 80 °C. Measurements were performed on four replicates per sample.

Ash content

Oven-dry paper samples ranging from 0.5 to 2.0 g were placed in the muffle furnace at 575 ± 25 °C for at least 4 h and dried to a constant weight. The ash content was determined according to Sluiter et al. (2008b) in a single replicate per sample.

Fibre length determination

Paper samples (approximately 0.1 g) were placed into polyethylene vials and allowed to swell with a small amount of water. Subsequently, they were pulped in a shaker. After complete shredding, the suspension was diluted several times so that in about 100 mL of suspension, there were at least one thousand fibres. The proportion of fibre length classes (< 0.5 mm, 0.5 to 1 mm, 1 to 2 mm, 2 to 3 mm, 3 to 7 mm) was determined using a fibre analyzer FS-100 (Kajaani Oy Electronics, Kajaani, Finland). Measurements were performed on a single replicate per sample, and the number of fibres within each population of replicate ranged from 16,138 to 20,112 cells.

Light microscopy of fibres

Fibre composition was analysed using a Nikon ECLIPSE 80i light microscope equipped with a Nikon DS-Fi1 digital camera (Nikon Instruments Europe, Amsterdam, Netherlands) on a small amount of stained fibres representing the tested sample, according to the standard ISO 9184 (1990). The fibres were stained by adding 2 to 3 drops of various colouring agents (Herzberg, Graff, and Loffton-Merritt).

Breaking length

Breaking length of paper samples was measured in accordance with STN 500340 (1998). For the calculation of breaking length, it was necessary to measure the basic weight of the paper samples according to the STN EN ISO 536 (50 0310) (1999). Measurements were performed on twenty replicates per sample (150 × 15 mm).

Brightness

Brightness of paper samples was assessed according to the STN ISO 3688 (50 0240) (1994) using a Fibre Leucometer PL 11 6565 ZP (Carl Zeiss, Jena, Germany). Measurements were performed on 20 replicates per sample.

Statistical analysis

Data were analysed using one-way analysis of variance, and Duncan’s multiple range tests were used for comparisons of means. The Pearson correlation coefficients were calculated for the examined trait-trait linkages. The relationships were considered significant if P < 0.05. Multivariate associations among 15 paper traits were analysed using a principal component analysis (PCA) in order to describe patterns of covariation among the examined traits.

RESULTS AND DISCUSSION

Changes in the pH values

The pH value of paper is a significant factor in determining its ageing stability. The measurements presented in Table 1 show that paper documents from the 18th and 19th centuries had values ranging from slightly acidic to neutral pH. Paper documents from the 20th century were characterized by acidic pH values, whereas paper documents from the 21st century possessed slightly alkaline pH values. Acidity promotes acid hydrolysis, which accelerates cellulose degradation in paper (Wilson and Parks 1983). Lignocellulosic materials with low pH values and high lignin content generally exhibit faster degradation than alkaline or neutral materials. Along with the use of raw materials possessing high lignin contents, the acidic sulphite process of pulp production and/or the process of acidic sizing also contribute to paper degradation. A low pH value is the most significant factor in the degradation of most endangered paper documents (Vizárová et al. 2012). Alkaline or neutral papers are more stable due to the suppression of acid hydrolysis with regards to the degree of degradation and the rate of ageing (Zappala 1991; Area and Cheradame 2011). Maršala et al. (2009) published pH values of papers from the 20th century and the early years of the 21st century. The most acidic documents (pH 3.9) originated between the years 1920 and 1960. The authors also discovered that the number of documents with neutral pH values (pH 7) gradually increased from the

1990s to the present.

Table 1. Principal Paper Characteristics for the Examined Paper Documents

Trait/Year 1719 1762 1784 1807 1839 1853
pHS 6.69 ± 0.07 d 6.58 ± 0.08 e 7.70 ± 0.07 b 6.94 ± 0.06 c 6.51 ± 0.03 ef 6.47 ± 0.02 f
pHE 6.46 ± 0.0.6 f 7.66 ± 0.04 c 7.23 ± 0.05 d 7.82 ± 0.03 b 7.24 ± 0.03 d 7.18 ± 0.04 e
FA (mg/g) 1.42 ± 0.02 g 3.75 ± 0.02 a 2.12 ± 0.04 c 2.22 ± 0.02 b 1.03 ± 0.03 h 0.43 ± 0.02 k
AA (mg/g) 2.41 ± 0.02 e 5.80 ± 0.07 a 4.68 ± 0.04 b 3.70 ± 0.03 d 3.84 ± 0.02 c 0.76 ± 0.01 j
XYL (%) 0.00 ± 0.00 e 0.00 ± 0.00 e 0.00 ± 0.00 e 0.00 ± 0.00 e 0.00 ± 0.00 e 0.00 ± 0.00 e
GAL (%) 1.14 ± 0.02 d 0.87 ± 0.01 h 1.16 ± 0.01 c 1.23 ± 0.01 b 1.24 ± 0.02 b 1.00 ± 0.01 f
ARA (%) 2.30 ± 0.01 a 1.93 ± 0.01 c 1.05 ± 0.01 i 1.11 ± 0.01 h 1.04 ± 0.02 i 1.57 ± 0.01 f
MAN (%) 2.61 ± 0.02 e 1.80 ± 0.01 i 2.23 ± 0.01 f 1.95 ± 0.01 h 2.20 ± 0.01 fg 2.19 ± 0.02 g
GLC (%) 88.83 ± 0.09 c 89.44 ± 0.07 b 88.61 ± 0.04c 89.33 ± 0.27 b 88.22 ± 0.21 d 91.04 ± 0.15 a
HOL (%) 94.88 ± 0.06 b 94.04 ± 0.06 c 93.06 ± 0.04e 93.63 ± 0.24 d 92.70 ± 0.18 f 95.79 ± 0.14 a
LIG (%) 1.30 ± 0.02 g 1.20 ± 0.02 h 1.00 ± 0.01 i 1.20 ± 0.02 h 2.68 ± 0.01 f 1.20 ± 0.01 h
BW (g m–2) 96.20 ± 0.39 a 56.80 ± 0.43 k 81.71 ± 0.24 f 69.64 ± 0.22 i 87.93 ± 0.24 b 86.67 ± 0.20 d
BL (km) 2.03 ± 0.40 ef 2.48 ± 0.38 d 1.90 ± 0.23 f 2.13 ± 0.20 ef 2.22 ± 0.23 e 2.96 ± 0.18 c
B (% MgO) 37.53 ± 4.02 g 52.70 ± 2.20bc 46.60 ± 1.91 f 45.07 ± 2.51 f 50.61 ± 2.71 cd 47.63 ± 1.35 ef
ASH (%) 4.51 2.98 3.51 2.99 2.78 2.08

Data represent the means SD. Mean values followed by the same letters, a–i, within the same row across examined paper documents, are not significantly different at p < 0.05.

pHS, pH of paper surface; pHE, pH of aqueous paper extract; FA, formic acid; AA, acetic acid; XYL, d-xylose; GAL, d-galactose; ARA, l-arabinose; MAN,

d-mannose; GLC, d-glucose; HOL, holocellulose; LIG, lignin; BW, basis weight; BL, breaking length; B, brightness; ASH, ash content

Table 1- continued. Principal Paper Characteristics for the Examined Paper Documents

Trait/Year 1859 1920 1923 1950 1956 2007
pHS 6.27 ± 0.01 g 4.31 ± 0.06 j 4.35 ± 0.03 j 4.58 ± 0.07 i 4.79 ± 0.04 h 7.79 ± 0.07 a
pHE 7.63 ± 0.02 c 5.55 ± 0.03 i 5.62 ± 0.04 h 5.76 ± 0.02 g 5.41 ± 0.01 j 7.96 ± 0.02 a
FA (mg/g) 0.69 ± 0.03 j 1.59 ± 0.03 f 1.38 ± 0.04 g 1.83 ± 0.03 d 1.65 ± 0.03 e 0.80 ± 0.02 i
AA (mg/g) 1.97 ± 0.02 f 1.98 ± 0.04 f 1.21 ± 0.04 i 1.82 ± 0.01 g 1.66 ± 0.02 h 1.81 ± 0.02 g
XYL (%) 0.00 ± 0.00 e 3.26 ± 0.01 c 4.18 ± 0.01 b 3.16 ± 0.02 d 3.24 ± 0.04 c 11.18 ± 0.02 a
GAL (%) 0.96 ± 0.01 g 1.10 ± 0.01 e 1.47 ± 0.02 a 0.00 ± 0.00 j 0.00 ± 0.00 i 1.00 ± 0.01 f
ARA (%) 1.58 ± 0.02 f 1.98 ± 0.01 b 1.92 ± 0.01 c 1.64 ± 0.01 e 1.70 ± 0.01 d 1.54 ± 0.02 g
MAN (%) 1.59 ± 0.02 j 5.50 ± 0.03 b 7.02 ± 0.02 a 2.85 ± 0.03 c 2.64 ± 0.03 d 1.79 ± 0.02 i
GLC (%) 87.27 ± 0.15 e 35.38 ± 0.20 j 41.24 ± 0.23 i 72.85 ± 0.09 g 75.16 ± 0.08 f 71.15 ± 0.35 h
HOL (%) 91.39 ± 0.16 g 47.22 ± 0.13 l 55.83 ± 0.22 k 80.51 ± 0.05 j 82.74 ± 0.12 i 86.65 ± 0.30 h
LIG (%) 5.17 ± 0.01 e 31.70 ± 0.05 b 32.64 ± 0.04 a 10.50 ± 0.02 c 5.30 ± 0.02 d 0.40 ± 0.02 j
BW (g m–2) 50.97 ± 0.17 l 87.40 ± 0.12 c 64.10 ± 0.13 j 83.80 ± 0.10 e 78.30 ± 0.10 h 79.89 ± 0.07 g
BL (km) 4.87 ± 0.18 b 1.17 ± 0.10 g 2.15 ± 0.12 ef 2.82 ± 0.08 c 2.77 ± 0.08 c 5.94 ± 0.08 a
B (% MgO) 54.03 ± 1.28 b 34.40 ± 2.88 h 35.34 ± 2.69gh 47.86 ± 0.92 ef 49.58 ± 1.14de 82.73 ± 0.16 a
ASH (%) 2.11 21.92 12.26 10.02 12.95 12.04

Data represent the means SD. Mean values followed by the same letters, a–i, within the same row across examined paper documents, are not significantly different at p < 0.05.

pHS, pH of paper surface; pHE, pH of aqueous paper extract; FA, formic acid; AA, acetic acid; XYL, d-xylose; GAL, d-galactose; ARA, l-arabinose; MAN,

d-mannose; GLC, d-glucose; HOL, holocellulose; LIG, lignin; BW, basis weight; BL, breaking length; B, brightness; ASH, ash content

Strlič et al. (2007) studied paper documents from 1870, which were composed of 70% cotton and 30% year-old wheat and straw. The authors found that the pHE of these paper samples was acidic (pH 4.7). This was also the case for paper samples from 1874, which consisted of 50% wood pulp and 50% sulphite pulp. An alkaline pH (pH 8.8) was recorded in paper samples originating from 2002, which were composed of 70% bleached kraft pulp and 30% bleached sulphite pulp.

Changes in the amount of acetic and formic acids

The greatest amounts of carboxylic acids were found in the oldest examined paper document from 1762 (Table 1), in which the content of acetic acid reached 3.8 mg g–1 and formic acid reached 5.8 mg g–1.

High values of carboxylic acids were also found in paper documents from 1807. Jablonsky et al. (2012b) measured 4.8 mg g–1 of acetic acid in paper documents that were subjected to an accelerated ageing period of 60 days. Formic acid penetrates deeper inside stacks of papers than acetic acid, which reveals its potential for degradation in archives and libraries (Tétreault et al. 2013). The removal of volatile organic compounds has a significant positive effect on paper stability and can, in certain cases, double their lifetime expectancy (Strlič et al. 2011). In addition, in the present work it was found that the contents of acetic and formic acid were mutually correlated (see supplementary Table S1).

Changes in contents of saccharides, lignin, and ash

A significant proportion of paper is composed of saccharides. Based on the quantitative results of neutral saccharides in paper documents from the 18th and 19th centuries, these samples contained the highest proportion of glucose, very small amounts of other hemicellulose components (i.e., galactose, mannose, and arabinose), and no xylose (Table 1).

The degradation of hemicellulose begins with deacetylation, forming acetic acid that catalyzes both the hydrolysis of glycosidic bonds in polysaccharides and the subsequent reactions of arising monosaccharides (Nuopponen et al. 2004; Esteves and Pereira 2009). Until 1859, the amount of glucose ranged from about 87 to 91%. In 1920, the amount of glucose dropped to 35.4% because wood pulp at the time was an important component of printing paper.

Paper documents from 1920 and 1923 were made from softwood pulp, as is evidenced by the high content of both lignin and glucomannan in their paper samples. These documents also had low values for breaking length and brightness.

The presence of lignin chromophores in the examined paper samples resulted in a significant yellowing of the paper. Paper documents from 1920 contained a high content of ash (21.9%), which may explain their low strength, e.g., in comparison to the paper document from the same period (1923), which contained a lower ash content. The 1923 document’s breaking length was almost twice as high. High filler content has an indirect effect on the surface strength and the stiffness of paper (Ciullo 1996).

Changes in fibre length distributions

The results of fibre length distributions (Fig. 1) showed that the largest proportion of fibres in the oldest examined paper documents (until 1859) were in the 0.5 to 1.0 mm length class (ranging from 38.5 to 46.0%).

Fig. 1. Proportion of fibre length classes across the examined paper documents

Fig. 2. Light microscopy images of fibres present in the examined paper documents. Some images are adapted from Gojný et al. (2014)

These paper samples also contained a relatively high proportion of fibres longer than 1 mm. The microscopic analysis of fibres (Fig. 2) revealed that the paper documents from the 18th and 19th centuries were primarily made from rag pulp (cotton, flax, hemp). According to Kilpinen (1994), shorter cotton fibres (3 to 5 mm) were used in the past to make paper. Microscopic analyses provided by Collings and Milner (1984) concerning paper specimens made in Europe between 1400 and 1800 primarily showed mixtures of hemp and flax fibres with higher concentrations of hemp (e.g., 75%) in earlier paper samples. While both cotton and cotton-containing fabrics were available in Europe during this period, they were not commonly used to generate substantial cotton-rag material for papermaking until the 19th century. Significant quantities of cotton fibre are therefore rare in paper documents before 1800.

Paper documents from the 20th century contained fibres within the 0.5 to 1.0 mm length class, from 28% (1950) to 40% (1956), with a large proportion of fibres longer than 2 mm (from 16% to 30%). The microscopic analysis of paper documents from 1920 and 1923 confirmed that these paper samples were made mostly from softwood pulp.

Paper documents from 1950 and 1956 consisted mainly of bleached softwood pulp, whereas paper samples from 2007 were produced from bleached hardwood pulp (Fig. 2). The latter paper documents contained the largest proportion of fibres at a length class of 0.5 to 1.0 mm (53.8%). Pulp fibre-length distribution that is observed after pulping and/or papermaking is a function of both the original fibre length distribution and the effects of processing. Because fibre length distributions affect tensile strength, tear, opacity, porosity, and many other paper properties, papermakers have developed many techniques to measure fibre length averages (Clark 1985). Other studies have also reported on the significant relationship between fibre length and the strength of paper (e.g., breaking length and tear index) (Seth and Page 1988; Niskanen 1998; Molteberg and Høibø 2006).

Correlated traits

Statistical analysis of the examined trait linkages revealed that some traits affected other characteristics, either positively or negatively. Correlation coefficients among the examined traits are presented in the supplementary Table S1.

Within the examined paper documents, there were significant relationships between the pH and the contents of mannose, glucose, total yield of saccharides (i.e., holocellulose), and lignin. Figure 3a shows the correlation of pHE on the amount of holocellulose, which peaked at a pHE of 7.18. The data was fitted with a second-order polynomial, and the measurements of the pHE explained 61% of the variation in the content of holocellulose (R2 = 0.610, P = 0.014).

Paper brightness was also related to the pHE (Fig. 3b), whereby the measurements of the pHE explained 36% of the variation in the brightness of the paper documents (R2 = 0.356, P = 0.040). Previous studies have shown that acidity accelerates the degradation of cellulose, holocellulose, and paper via acid-catalyzed hydrolysis (Wilson and Parks 1983). Weak, discoloured, and brittle paper correlated with low pH, while neutral and alkaline papers were generally in a much better state of preservation (Sobucki and Drewniewska-Idziak 2003).

Furthermore, there was a positive relationship between the breaking length and the proportion of fibre length class 0.5 to 1.0 mm (Fig. 3c). Measurements of the proportion of fibre length class 0.5 to 1.0 mm explained 53% of the variation in the breaking length of paper documents (R2 = 0.529, P = 0.007).

Fig. 3. Relationships between the pH of aqueous paper extract and the content of holocellulose (a) and to paper brightness (b); the relationship of the proportion of fibre length class 0.5 to 1.0 mm to the breaking length of paper documents (c)

Greater fibre length and higher fibre strength supports higher paper strength (Fišerová et al. 2009). In this study, the fibre length class 0.5 to 1.0 mm was clearly the predominant proportion for the majority of the examined paper documents, and was thus primarily responsible for the strength of the paper samples. In addition, the amount of mannose negatively influenced the brightness of the paper because lignin, the major lignocellulosic component responsible for reducing brightness (Chen et al. 2012), correlated well with the mannose content (see supplementary Table S1).

Fig. 4. Positions of 15 paper traits on the first and second axes of the principal component analysis (PCA). The bottom and left-hand axes refer to the paper traits, whereas the top and right-hand axes refer to the printing year for the examined paper documents.

Associations among paper traits

A principle component analysis (PCA) was conducted to evaluate how the examined traits were associated (Fig. 4). The first axis explained 46% of the variation and showed strong positive loadings for the contents of holocellulose and glucose, as well as for the pHS and pHE. The negative side of the axis indicated strong loadings for the contents of lignin, mannose, and ash.

The second axis explained 20% of the variation and showed strong positive loadings for the breaking length, brightness, and xylose content. The negative side of the axis indicated strong loadings for the contents of both formic and acetic acids.

In addition, four groups of compact homogeneous clusters could be distinguished in a multivariate paper trait analysis that were clearly segregated from each other. Paper documents printed in 1762, 1784, 1807, and 1839 formed the first cluster, followed by clusters of paper documents printed in 1853 and 1859, then 1920 and 1923, and finally 1950 and 1956. The paper sample printed in 1719 formed a single segregated specimen; the same case was also observed for the paper document originating from 2007. The latter paper document was the most segregated specimen from all of the other groups, and traits such as breaking length and brightness were closely associated solely with this document.

CONCLUSIONS

  1. Paper documents originating from the 18th and 19th centuries were characterized by slightly acidic to neutral pH values, those from the 20th century by acidic values, and those from 2007 by slightly alkaline values. The contents of mannose, glucose, holocellulose, and lignin were significantly affected by the pH value of paper.
  2. Paper documents from the 18th and 19th centuries were made from non-wood fibres (rag), those from the 1920s from softwood pulp, those from the mid of the 20th century from softwood bleached pulp, and lastly those from 2007 from bleached hardwood pulp.
  3. Breaking length was influenced by both the fibre length class of 0.5 to 1.0 mm, and the degree of delignification as represented by brightness.
  4. The lowest strength value was associated with 1920 paper documents made from wood fibres and containing very high amounts of both lignin and ash (fillers). Paper documents from 1920 and 1923 also had the lowest percentage of brightness. Therefore, paper documents originating from this period deserve very special attention and protection in public archives and libraries.
  5. The paper document printed in 2007 was the most segregated specimen from all of the other groups of paper document clusters with regards to a multivariate paper trait analysis.
  6. To make the paper more resistant to aging it is crucial to keep alkaline pH values, lower the air humidity and the temperature of the environment where the books are kept.

ACKNOWLEDGMENTS

The authors thank Mrs. E. Ritch-Krč for language revision. This publication is the result of the Project “Raising Human Resources Capacity Towards the Transfer of Biomass Production and Processing Research and Development Results into Practice” (ITMS: 262110230087), supported by the Operational Programme “Education” funded by the European Social Fund (50%) and by the Slovak Scientific Grant Agency VEGA under contract No. 1/0521/15 (50%).

REFERENCES CITED

Amidon, T. E. (1981). “Effect of the wood properties of hardwood on kraft paper properties,” Tappi 64, 123-126.

Area, M. C., and Cheradame, H. (2011). “Paper aging and degradation: Recent findings and research methods,” BioResources 6(4), 5307-5337. DOI: 10.15376/biores.6.4.5307-5337

Begin, P., Deschatelets, S., Grattan, D., Gurnagul, N., Iraci, J., Kaminska, E., Woods, D. and Zou, X. (1998). “The impact of lignin on paper permanence: A comprehensive study of the aging behaviour of handsheets and commercial paper samples,” Restaurator 19(3), 135-154.

Chen, Y., Fan, Y., Tshabalala, M. A., Stark, N. M., Gao, J., and Liu, R. (2012). “Optical property analysis of thermally and photolytically aged Eucalyptus camaldulensis chemithermomechanical pulp (CTMP),” BioResources 7(2), 1474-1487. DOI: 10.15376/biores.7.2.1474-1487

Ciullo, P. A. (1996). Industrial Minerals and their Uses: A Handbook and Formulary, Noyes Publication, Westwood, NJ, USA.

Clark, J. D. A. (1985). “Fibrillation and fiber bonding,” in: Pulp Technology and Treatment for Paper, J. D. A. Clark (ed.), Miller Freeman Publications, San Francisco, CA, USA, pp. 160-180.

Collings, T., and Milner, D. (1984). “The nature and identification of cotton paper-making fibres in paper,” Paper Conservator 8(1), 59-71. DOI: 10.1080/03094227.1984.9638458

Dinwoodie, J. M. (1965). “The relationship between fibre morphology and paper properties of: A review of literature,” Tappi 48, 440-447.

Dupont, A. L. (2003). Gelatine Sizing of Paper and its Impact on the Degradation of Cellulose during Ageing, Ph.D. Dissertation, University of Amsterdam, Amsterdam, Netherlands.

Esteves, B., and Pereira, H. (2009). “Wood modification by heat treatment: A review,” BioResources 4(1), 370-404. DOI: 10.15376/biores.1.1.1-2

Fišerová, M., Gigac, J., and Balberčák, J. (2009). “Relationship between fibre characteristics and tensile strength of hardwood and softwood kraft pulps,” Cellulose Chemistry and Technology 44(7-8), 249-253.

Gandini, A., and Pasquini, D. (2012). “The impact of cellulose fibre surface modification on some physico-chemical properties of the ensuing papers,” Industrial Crops and Products 35(1), 15-21. DOI: 10.1016/j.indcrop.2011.06.015

Gojný, J., Češek, B., Mikala, O., and Čabalová, I. (2014). “Fibres length distribution of historical paper documents,” Acta Facultatis Xylologiae 56(2), 55-61. (in Slovak)

Hubbe, M. A. (2005). “Acidic and alkaline sizings for printing, writing, and drawing papers,” The Book and Paper Group Annual 23, 139-151.

ISO 9184 (1990). “Paper, board and pulps – Fibre furnish analysis,” International Organization for Standardization, Geneva, Switzerland.

Jablonsky, M., Botkova, M., and Hrobonova, K. (2012a). “Accelerated ageing of wood-containing papers: Formation of weak acids and deterioration of tensile strength,” Wood Research 57, 3, 419-434.

Jablonsky, M., Hrobonova, K., Katuscak, S., Lehotay, J., and Botkova, M. (2012b). “Formation of acetic and formic acid in unmodified and modified papers during accelarated ageing,” Cellulose Chemistry and Technology 46(5-6), 331-340.

Kilpinen, O. (1994). “Nonwood specialty pulps,” in: Nonwood Plant Fiber Pulping – Progress Report 21, TAPPI Press, Atlanta, GA, pp. 9-18.

Kraková, L., Chovanová, K., Selim, S. A., Šimonovičová, A., and Puškarová, A. (2012). “A multiphasic approach for investigation of the microbial diversity and its biodegradative abilities in historical paper and parchment documents,” International Biodeterioration and Biodegradation 70, 117-125. DOI: 10.1016/j.ibiod.2012.01.011

Lichtblau, D., Strlič, M., Trafela, T., Kolar, J., and Anders, M. (2008). “Determination of mechanical properties of historical paper based on NIR spectroscopy and chemometrics – A new instrument,” Applied Physics A. 92, 191-195. DOI: 10.1007/s00339-008-4479-1

Maršala, M., Kuka, I., Bukovský, V., and Švehlová, D. (2009). “The most important parameters to the assemment the durability of paper documents,” Knižnica 10, 35-38 (in Slovak).

Mohlin, U. B., Dahlbom, J., and Hornatowska, J. (1996). “Fibre deformation and sheet strength,” Tappi Journal 79(6), 105-111.

Molteberg, D., and Høibø, O. (2006). “Development and variation of wood density, kraft pulp yeild and fibre dimension in young Norway spruce (Picea abies),” Wood Science and Technology 40, 173-189. DOI: 10.1007/s00226-005-0020-2

Niskanen, K. (1998). Paper Physics, Fapet Oy, Helsinki, Finland.

Nuopponen, M., Vuorinen, T., Jamsä, S., and Viitaniemi, P. (2004). “Thermal modifications in softwood studied by FT-IR and UV resonance Raman spectroscopies,” Journal of Wood Chemistry and Technology 24(1), 13-26. DOI: 10.1081/WCT-120035941

Paavilainen, L. (1993a). “Conformability − flexibility and collapsibility − of sulphate pulp fibres,” Paperi ja Puu 75(9), 689-702.

Paavilainen, L. (1993b). “Importance of cross-dimensional fibre properties and coarseness for the characterisation of softwood sulphate pulp,” Paperi ja Puu 75(5), 343-351.

Page, D. H, and Seth, R. (1979). “The extensional behaviour of commercial mechanical pulps,” Pulp and Paper Canada 80(8), T235-T237.

Polovka, M., Polovkova, J., Vizarova, K., Kirschnerova, S., Bielikova, L., and Vrska, M. (2006). “The application of FTIR spectroscopy on characterization of paper samples, modified by Bookkeeper process,” Vibrational Spectroscopy 41(1), 112-117. DOI: 10.1016/j.vibspec.2006.01.010

Seth, R. S. (1990). “Fibre quality factors in papermaking – I. The importance of fibre length and strength,” in: Materials Interactions Relevant to the Pulp, Paper and Wood Industries, D. F. Caulfield, J. D. Passaretti, and S. F. Sobczynski (Eds.), Materials Research Society, San Francisco, CA, USA, pp. 125-142.

Seth, R. S., and Page, D. H. (1988). “Fibre properties and tearing resistance,” Tappi Journal 71, 103-107.

Schmidt, J. A., Rye, C. S., and Gurnagul, N. (1995). “Lignin inhibits autoxidative degradation of cellulose,” Polymer Degradation and Stability 49(2), 291-297.

Sluiter, A., Hames, B., Ruiz, R., Scarlata, C., Sluiter, J., and Templeton, D. (2008a). Determination of Sugars, Byproducts, and Degradation Products in Liquid Fraction Process Samples (NREL/TP-510-42623), National Renewable Energy Laboratory, Golden, CO, USA.

Sluiter, A., Hames, B., Ruiz, R., Scarlata, C., Sluiter, J., and Templeton, D. (2008b). Determination of Ash in Biomass (NREL/TP-510-42622), National Renewable Energy Laboratory, Golden, CO, USA.

Sluiter, A., Hames, B., Ruiz, R., Scarlata, C., Sluiter, J., Templeton, D., Crocker, D. (2010). Determination of Structural Carbohydrates and Lignin in Biomass (NREL/TP-510-42618), National Renewable Energy Laboratory, Golden, CO, USA.

Sobucki, W., and Drewniewska-Idziak, B. (2003). “Survey of the preservation status of the 19th and 20th century collections at the National Library in Warsaw,” Restaurator 24(3), 189-201. DOI: 10.1515/REST.2003.189

STN 500340 (1998). “Paper and board. Determination of tensile properties,” VÚPC, Bratislava, Slovakia.

STN 500374 (1996). “Testing of pulp and paper. Surface pH measurement of paper and pulp,” VÚPC, Bratislava, Slovakia.

STN EN ISO 536 (50 0310) (1999). “Paper and board. Determination of grammage,” VÚPC, Bratislava, Slovakia.

STN ISO 3688 (50 0240) (1994). “Pulps. Measurement of diffuse blue reflectance factor (ISO brightness),” VÚPC, Bratislava, Slovakia.

STN ISO 6588 (50 0381) (1993). “Paper, board and pulps. Determination of pH of aqueous extracts,” VÚPC, Bratislava, Slovakia.

Strlič, M., Cigić, I. K., Kolar, J., de Bruin, G., and Pihlar, B. (2007). “Non-destructive evaluation of historical paper based on pH estimation from VOC emissions,” Sensors 7(12), 3136-3145. DOI: 10.3390/s7123136

Strlič, M., Cigić, I. K., Možir, A., de Bruin, G., Kolar, J., and Cassar, M. (2011). “The effect of volatile organic compounds and hypoxia on paper degradation,” Polymer Degradation and Stability 96(47), 608-615. DOI: 10.1016/j.polymdegradstab.2010.12.017

Strlič, M., Kolar, J., Kočar, D., Drnovšek, T., Selih, V., Susič, R., and Pihlar, B. (2004). “What is the pH of alkaline paper?,” e-Preservation Science 1, 35-47.

Tamolang, F. N., Wangaard, F. F., and Kellogg, R. M. (1968). “Hardwood fibre strength and pulp-sheet properties,” Tappi Jounal 51, 19-27.

Tétreault, J., Dupont, A. L., Bégin, P., and Paris, S. (2013). “The impact of volatile compounds released by paper on cellulose degradation in ambient hygrothermal conditions,” Polymer Degradation and Stability 98(9), 1827-1837. DOI: 10.1016/j.polymdegradstab.2013.05.017

Trafela, T., Strlič, M., Kolar, J., Lichtblau, D. A., Anders, M., Mencigar, D. P., and Pihlar, B. (2007). “Nondestructive analysis and dating of historical paper based on IR spectroscopy and chemometric data evaluation,” Analytical Chemistry 79(16), 6319-6323. DOI: 10.1021/ac070392t

Vainio, A. (2007). Interfibre Bonding and Fibre Segment Activation in Paper – Observations on the Phenomena and their Influence on Paper Strength Properties, Ph.D. Dissertation, Helsinki University of Technology, Helsinki, Finland.

Ververis, C., Georghiou, K., Christodoulakis, N., Santas, P., and Santas, R. (2004). “Fiber dimensions, lignin and cellulose content of various plant materials and their suitability for paper production,” Industrial Crops and Products 19(3), 245-254. DOI: 10.1016/j.indcrop.2003.10.006

Vizárová, K., Kirschnerová, S., Kačík, F., Briškárová, A., Šutý, Š. and Katuščák, S. (2012). “Relation between the decrease of degree of polymerisation of cellulose and the loss of groundwood pulp paper mechanical properties during accelerated ageing,” Chemical Papers 66(12), 1124-1129.

Wangaard, F. F., and Williams, D. L. (1970). “Fibre length and fibre strength in relation to tearing resistance of hadwood pulp,” Tappi Journal 53, 2153-2154.

Whitmore, P. M., and Bogaard, J. (1994). “Determination of the cellulose scission route in the hydrolytic and oxidative degradation of paper,” Restaurator 15(1), 26-45.

Wilson, W. K, and Parks, E. J. (1983). “Historical survey of research at the National Bureau of Standards on materials for archival records,” Restaurator 5(3-4), 191-241. DOI: 10.1515/rest.1983.5.3-4.191

Zappala, A. (1991). “An international survey on standardizing art papers and others intended for conservation,” Restaurator 12(1), 18-35. DOI: 10.1515/rest.1991.12.1.18

Zervos, S. (2007). “Accelerated ageing kinetics of pure cellulose paper after washing, alkalization and impregnation with methylcellulose,” Restaurator 28(1), 55-69. DOI: 10.1515/rest.2007.55

Zervos, S. (2010). “Natural and accelerated ageing of cellulose and paper: A literature review, “ in: Cellulose: Structure and Properties, Derivatives and Industrial Uses, A. Lejeune, and T. Deprez (Eds.), New York: Nova Publishing, pp. 155-203.

Zotti, M., Ferroni, A., and Calvini, P. (2008). “Microfungal biodeterioration of historic paper: Preliminary FTIR and microbiological analyses,” International Biodeterioration and Biodegradation 62(2), 186-194. DOI: 10.1016/j.ibiod.2008.01.005

Zou, X., Uesaka, T., and Gurnagul, N. (1996a). “Prediction of paper permanence by accelerated aging I. Kinetic analysis of the aging process, Cellulose 3, 243-267. DOI: 10.1007/BF02228805

Zou, X., Uesaka, T., and Gurnagul, N. (1996b). “Prediction of paper permanence by accelerated aging II. Comparison of the predictions with natural aging results,” Cellulose 3, 269-279. DOI: 10.1007/BF02228806

Zou, X., Gurnagul, N., and Uesaka, T. (1993). “The role of lignin in the mechanical permanence of paper. Part I: Effect of lignin content,” Journal of Pulp and Paper Science 19(6), j235-j239.

Article submitted: February 25, 2016; Peer review completed: October 9, 2016; Revised version received: January 24, 2017; Tentative acceptance: February 7, 2017; Published: February 21, 2017.

DOI: 10.15376/biores.12.2.2618-2634

Table S1.

Pearson Correlation Coefficients for the Examined Paper Traits

Trait pHS pHE FA AA XYL GAL ARA MAN GLC HOL LIG BW BL B ASH FL1 FL2 FL3 FL4 FL5
pHS 0.884 0.007 0.447 –0.062 0.372 –0.458 –0.727 0.712 0.766 –0.792 0.018 0.377 0.559 –0.651 –0.128 0.704 0.076 –0.615 –0.591
pHE 0.884 0.037 0.477 –0.103 0.397 –0.489 –0.717 0.666 0.703 –0.695 –0.312 0.484 0.597 –0.690 –0.060 0.728 –0.035 –0.587 –0.655
FA 0.007 0.037 0.772 –0.261 –0.130 0.093 –0.056 0.096 0.054 –0.061 –0.306 –0.427 –0.178 –0.067 0.051 –0.322 0.338 0.077 0.051
AA 0.447 0.477 0.772 –0.414 0.194 –0.325 –0.388 0.441 0.401 –0.393 –0.225 –0.284 0.038 –0.436 0.165 0.078 0.279 –0.291 –0.467
XYL –0.062 –0.103 –0.261 –0.414 –0.141 0.140 0.217 –0.509 –0.382 0.229 0.078 0.562 0.608 –0.623 –0.615 0.266 –0.116 0.278 0.382
GAL 0.372 0.397 –0.130 0.194 –0.141 –0.109 0.253 –0.085 –0.079 0.170 –0.073 –0.156 –0.169 –0.210 0.255 0.268 0.051 –0.262 –0.762
ARA –0.458 –0.489 0.093 –0.325 0.140 –0.109 0.424 –0.408 –0.396 0.407 0.033 –0.096 –0.301 0.407 0.270 –0.271 –0.429 0.273 0.294
MAN –0.727 –0.717 –0.056 –0.388 0.217 0.253 0.424 –0.900 –0.913 0.957 0.040 –0.452 –0.584 0.676 0.047 –0.544 0.128 0.548 0.232
GLC 0.712 0.666 0.096 0.441 –0.509 –0.085 –0.408 –0.900 0.988 –0.945 –0.030 0.205 0.295 –0.899 0.055 0.441 0.044 –0.619 –0.338
HOL 0.766 0.703 0.054 0.401 –0.382 –0.079 –0.396 –0.913 0.988 –0.972 –0.017 0.308 0.407 –0.876 –0.043 0.528 0.033 –0.629 –0.317
LIG –0.792 –0.695 –0.061 –0.393 0.229 0.170 0.407 0.957 –0.945 –0.972 –0.059 –0.373 –0.533 0.745 0.111 –0.571 0.015 0.614 0.239
BW 0.018 –0.312 –0.306 –0.225 0.078 –0.073 0.033 0.040 –0.030 –0.017 –0.059 –0.326 –0.153 0.222 –0.172 –0.225 0.147 0.274 0.263
BL 0.377 0.484 –0.427 –0.284 0.562 –0.156 –0.096 –0.452 0.205 0.308 –0.373 –0.326 0.870 –0.124 –0.348 0.727 –0.395 –0.279 –0.041
B 0.559 0.597 –0.178 0.038 0.608 –0.169 –0.301 –0.584 0.295 0.407 –0.533 –0.153 0.870 –0.121 –0.492 0.770 –0.211 –0.305 –0.071
ASH –0.651 –0.690 –0.067 –0.436 –0.623 –0.210 0.407 0.676 –0.899 –0.876 0.745 0.222 –0.124 –0.121 –0.107 –0.400 –0.156 0.590 0.521
FL1 –0.128 –0.060 0.051 0.165 –0.615 0.255 0.270 0.047 0.055 –0.043 0.111 –0.172 –0.348 –0.492 –0.107 –0.029 –0.590 –0.417 –0.507
FL2 0.704 0.728 –0.322 0.078 0.266 0.268 –0.271 –0.544 0.441 0.528 –0.571 –0.225 0.727 0.770 –0.400 –0.029 –0.432 –0.771 –0.554
FL3 0.076 –0.035 0.338 0.279 –0.116 0.051 –0.429 0.128 0.044 0.033 0.015 0.147 –0.395 –0.211 –0.156 –0.590 –0.432 0.400 0.215
FL4 –0.615 –0.587 0.077 –0.291 0.278 –0.262 0.273 0.548 –0.619 –0.629 0.614 0.274 –0.279 –0.305 0.590 –0.417 –0.771 0.400 0.661
FL5 –0.591 –0.655 0.051 –0.467 0.382 –0.762 0.294 0.232 –0.338 –0.317 0.239 0.263 –0.041 –0.071 0.521 –0.507 –0.554 0.215 0.661

Note: Correlation coefficients presented in bold are significant at P < 0.05.

pHS, pH of paper surface; pHE, pH of aqueous paper extract; FA, formic acid; AA, acetic acid; XYL, d-xylose; GAL, d-galactose; ARA, l-arabinose; MAN, d-mannose; GLC, d-glucose; HOL, holocellulose; LIG, lignin; BW, basis weight; BL, breaking length; B, brightness; ASH, ash content; FL1, fibre length class <0.5 mm; FL2, fibre length class 0.5–1 mm; FL3, fibre length class 1–2 mm; FL4, fibre length class 2–3 mm; FL5, fibre length class 3–7 mm.