1 3th Triennial Meeting Rio de Janeiro
22-27 September2002
ICOM COMMITTEE FOR C O N S E R V A T I O N
COMITÉ DE L'ICOM POUR LA CONSERVATION IC O M COMIT É PARA L A C O NSERVACIÓN
PREPRINTS Volume II
Abstract
It is well known that paper is not stable in time. There are several causes of paper decay, but acidification is by far the main reason. In general the pH is used to express paper acidity. Tl;e determination of the pH of paper, a solid material, is a difficult task. Several methods exist, nine of which were evaluated in this study. All methods are destructive and can cause serious damage to the paper investigated. To limit the damage a micro-extraction method was developed that only requires about 40 ug paper for one pH measurement.
A different way to express the acidity of a paper sample is the alkaline reserve, a parameter that is often used when analyzing paper with a pH higher than seven. Its determination is quite time- and product-consuming and labour-intensive. This work tested if the pH can be an indicator for the alkaline reserve.
Keywords
acidity/pH of paper, comparison of pH determining methods, micro-destructive sampling, alkali reserve
The acidity of paper. Evaluation of methods to measure the pH of paper samples
Steven Saverwyns*, Valerie Sizaire and Jan Wouters
Royal Institute for Cultural Heritage
Jubelpark 1
B-1000 Brussels
Belgium
Fax: + 32 (0)2 732 01 05
E-mail: , ,
Web site:
Introduction
Of all writing and drawing materials that people have employed down the ages, like waxed boards, leaves, bronze, silk, and clay tablets, paper is the most widely used around the world and has become one of the most important bearers of our history and culture.
Unfortunately, as with organic materials in general, paper is susceptible to several decay processes that may lead to irreversible losses in relatively short times (McCrady 1996, Havermans 1997, Begin 1999, Bukovsky 2000). One of the most important factors that accelerate the decay of paper is acidity, usually expressed by the pH of an aqueous extract of the paper. Acidic papers yield extracts with a low pH, and hence with a high proton concentration. Those papers can become brittle and, in some extreme cases, pulverize. At the molecular level this means that the long cellulose chains present in a strong or 'healthy' paper, are broken down by hydrolysis in the presence of water bound to the paper (Whitmore 1994, El-Saied 1998, Neevel 1999, Dupont 2000). The speed of this hydrolytic reaction increases significantly when the proton concentration in the molecular environment of cellulose is high.
By definition, the pH is measured in solutions, making the pH determination of paper, a solid material, a difficult task. In most cases this problem is dealt with by taking a certain amount of paper and adding water, so that a fixed ratio paper/ water is obtained, followed by measuring the pH of the aqueous extract. To assure that the pH values found are reproducible, the procedure of taking the sample and measuring the pH is described in standards. Unfortunately for the conservation community, these standards stem from the paper industry, requiring between 1 and 2 grams of paper. Besides the methods described in standards other possibilities exist to determine the pH of paper, and many of those are frequently used by conservators.
The goal of this research was to compare existing methods for the pH determination of paper, and to develop a micro-destructive method allowing conservators to use an extraction technique warranting accurate pH readings.
Materials and methods
pH determination tests
Nine different methods for the determination of the pH of paper were evaluated: two cold extractions, two hot extractions, four different indicators (Abbey pen, pHydrion pencil, bleeding and non-bleeding indicator paper) and a surface electrode.
extractions
Two standards, ISO 6588 (1981) and ASTM D 778 (1998), both describing a ho
and a cold extraction, were examined.
For the ISO cold extraction 2 g of paper is cut in pieces with maximum dimensions of 0.5 x 0.5 cm. 100 mL distilled water is added and the mixture is left to stand for one hour, but should at least be shaken once. The pH is measured in the extract, after removal of the paper shred, using a conventional pH electrode (in this work: Hamilton Minitrode).
^Author to whom correspondence should be addressed
Graphic documents
The ISO hot extraction is analogous to the cold extraction except that the extraction is performed using boiling distilled water. After rapid cooling of the extract, and removal of the shred, the pi I is measured.
The ASTM cold extraction method consists of" the extraction of a 1 g test specimen in 70 inL of cold distilled water (25 ±5°C") for one hour, and determination of pH without filtration using a coininerci.il pi I meter.
The same procedure is followed for the ASTM hot extraction except that the test specimen is boiled for one hour. After cooling down of the extract the pH is measured. For both the ASTM methods, pure N , was passed through the solutions until the pH is measured to prevent absorption of CO2.
indicators
All indicator tests are based on the development of pi I-dcpendcnt hues and a scale of reference colours. The simplest example is the Abbey pen. A line is drawn on a dry paper. It the line colours yellow, the paper is acidic (or has a pH below 7); il the colour remains purple, the paper is neutral or alkaline. The other indicator tests allow determination ot a pH value, rather than indicate it a paper is acidic or not. For the bleeding and non-bleeding indicator strips as well as for the pi lydrion pencil the paper needs to be wetted, which can leave stains on the paper. Also the indicators themselves leave stains, with the exception ot the non-bleeding ones.
Surface Electrode
This test was performed according to the Tappi 'T 52V om-99' standard (1999): a flat combination electrode is immersed in a drop ol water on the surface of the paper sample. The paper sample is backed with a non-absorbing, resilient material such as gum rubber. After about ten minutes' immersion time the pH can be read on the pH meter.
Preparation of test papers
Before the different methods can be tested on their accuracy, and compared to each other, papers with known pH are necessary. The easiest way to obtain such papers is to treat neutral papers with buffers with known pH. For the preparation ot acidic and neutral papers citric acid was used, tor alkaline papers trishydroxymethylaminomethane (or 'tris'), both in a concentration of 250mmol/L. The pH of the buffers was adjusted to the desired value with a sodium hydroxide solution. Pure cellulose paper (Whatman, chromatography paper 1 CHR, 25 x 25 cm) was tised. A sheet was placet! in a small tank and the buffer was added. The tank was shaken gently, and after five minutes the paper was removed. It was placed between two Hollytex sheets and left to dry for 24 hours. The first hour it was turned over every 15 minutes to prevent migration of the buffer.
I hiring any extraction step thepH will change as a consequence of dilution. The factor of dilution must be taken into account to correct expected pH readings and to make proper evaluations of accuracy. This factor was determined using neutral Whatman paper treated with a phosphate solution according to the buffering procedure previously described. This was followed by an extraction as described in the ISC) standard. With the aid of an absorbance spectrophotometer the phosphate concentration was determined in the original phosphate solution and in the extract. From these results the dilution factor was calculated to be 30. This implies that the pH of the paper extracts needs to be compared with the pH of the 30-fold diluted buffer solutions.
Refinement ofthe Tappi 'T 529 om-99 standard
The Tappi 'T 529 om-99' standard does not describe how much water needs to be added to the paper before the surface electrode is placed on the paper, nor are the dimensions of the paper sample mentioned. Moreover, the diffusion ot the water cannot be accounted for. However, it seemed reasonable to us to measure surface pHs at known paper/liquid ratios. To this end surface pFIs were measured on cut paper samples (buffered at pH 3.64) ofknown dimensions (either 1.1 x 1.1 cm or 2.3 x 2.3 cm) and using 50 u.L of water in each case. pH readings differed significantly and only in thecase of the smaller samples were accurate pH-values found (see Figure la and b). The higher pH for the larger sample can be explained by the fact that the pH-determining components are highly water soluble, and that the papers easily absorb the water. This results in a migration of the buffer components to the edges of the paper sample. When the paper sample exceeds the dimensions of the electrode surface, the buffer concentration underneath the electrode will be lower than expected, and in the case of acidic papers a higher pH will be found.
The amount of water added also influenced the pH measured. This is illustrated in Figure 2. When only 10 or 20 μL of water was added to a paper sample (dimensions 1.1 x 1.1 cm), too high or too low a pH was found, for an acidic or an alkaline paper respectively. Volumes larger than 50 μT on the other hand resulted in small deviations, probably due to dilution effects. From these results it was concluded that a volume of 40 μL is desired to obtain accurate pH values.
These results indicate that small alterations to the Tappi Standard are advisable. According to the dimensions of the paper sample, or the volume of the water droplet, in practice the paper wettability, results can vary widely.
Results and discussion
Comparative studies
accuracy tests with Reference Papers
To test the accuracy of the different pH methods previously described, papers (coded W 1, W2, W3 and W4) buffered at 4 different pi 1 values were used: 3.40, 4.94, 6.48 and 8.50 respectively. These pH values were corrected for dilutions as described in 'Materials and methods' (above). From Table 1 it is clear that .ill extraction methods lead to accurate pH values. Results found with the surface electrode must be compared with the pH values of the non-diluted buffers since in this case almost no dilution effects occur when measuring the pH. Therefore we must conclude that also with the surface electrode acceptable results are obtained (when measured under controlled conditions of paper dimensions and volume of water added). Also the pH values of the indicators need to be compared to the values of the non-diluted buffer solutions. Predictions with the Abbey pen were correct, but offer very limited information concerning the true pH. The pH values found with the pFIydriou pencil and with the indicator papers can in some cases diverge Ironi the butter pM with mure than 0.5 pH unit, illustrating that these tests are not very accurate and only lead to a rough estimation of the pH. Besides this, the interpretation of the colour of the indicators is liable to errors. These results agree with earlier studies (Brandis 1993).
In general it can be concluded that only the extraction techniques and the surface electrode lead in all cases to accurate results. The errors on the measurements, which are a measure of the precision, are significantly higher with the surface electrode as compared to extracts. Therefore extraction techniques should be preferred.
Tests and aged papers
Once the accuracy and the precision of the different methods were established using the buffered papers, some natural and artificially aged papers were also analyzed (see Fable 2). No conclusions concerning the accuracy could be drawn (since the 'true' pH values are not known). When we focus on the extraction techniques, it is clear that tor the acidic papers in almost all cases the hot extractions cause lower pH values than the cold extractions, and higher pH values for alkaline
Figure 1. pH measurements with a surface electrode, pHreadings as a function of time for a buffered paper withdimensions of 1. 1 x 1.1 cm (a) and 2.3 x 2.3cm (b). The rectangle selectsreadings used for calculations of means anddeviations.
Figure 2. pH measurements with a surface electrode. pH readings as a function of the volume of water added to the paper for an alkaline (x) and an acidic () paper.
Table 1. pH-values obtained with the different methods for buffered papers with known pH. The standard deviation {calculated for n = 2measurements) is given in bracket.s
Table 2. pH- values obtained with the different methods for aged papers. The standard deviation ( calculated for n = 2 measurements ) is given in brackets.
Table 3. Comparison of pH- values obtained with the classiccoldextraction method and with the new miniaturized attraction method. T he standard deviation ismentioned in
brackets (n = 2, except for NP 1957, where n = 1)
Figure 3. Stabilization curves, showing the change in pH as a function of time, of some selected aged papers, when measuring with the surface electrode. The squares indicate the best moment for pH readings.
papers. The reason is the more efficient extraction ot pi [-determining compounds in the case ofhot extractions. Differences between the cold extractions as a group and between the hot extractions as a group are acceptable. The difference of 0.5 pH unit between the cold extractions for the alkaline paper can be attributed to the absorbance of CO2, leading to a lower pH value, when the extract is not flushed with N2, as is the case when using the ISC) standard.
pH values obtained with the surface electrode are always lower than the values of the cold extraction, again ascribed to the fact that little dilution occurs. Besides the worse precision obtained with the surface electrode, an additional disadvantage was noticed. The time to obtain a stable pH reading depends strongly on the paper sample as is illustrated in Figure 3. Some papers (for example, sample A2) do not easily absorb the water, resulting in long stabilization times. Others absorb the water readily, resulting in quick responses, but after some time the wetted paper dries out and the signal starts drifting (for example, sample A3). This implies that the pH needs to be followed as a function ot time to know when a stable pH reading can be clone. This is very time-consuming and labour-intensive.
For the indicators some problems in interpreting the colour-developed occurred when the paper samples were already coloured (tor example, yellow due to ageing effects). The Abbey pen still allows the prediction of an acidic pFI or not, while for the other indicators again, deviating values from the ones measured with the extraction techniques are obtained.
Miniaturization ofthe ISO-cold extraction standard
Although the extractions lead to excellent results, the methods cannot be used on precious papers since they require too much sample to be taken. In order to make
Figure 4. Schematic presentation of asheet of abook showing the places the samples were taken for pH determination either with the miniaturized method ( nos 1-10 ) or with the classic ISO extraction method ( Strip 1-Strip 3 ).
Table 4. pH of paper as afunction of the distance from the edge of asheet, illustrating the high area specificity of the method. For three paper strips, all of which enclose several microdrill positions, the pH determined using the classic ISO extraction method is also given.
the extractions more accessible to the conservation community, a miniaturization procedure was worked out. The samples were taken using a so-called imcroilnll technique (Puchinger 2001, Wouters 2001), and weighed only about 40μg. This is a reduction in the amount of paper by a factor of 50,000. A volume ot water. with respect to the ratio paper/water mentioned in the ISC) standard, was added and the pi I ot the extract measured using a micro pH-electrode. Again the accuracy and the precision were tested using the buffered papers (W1-W4) and even on this scale results were excellent (Table 3). Only the alkaline paper shows a small deviating result.
Also tor some aged papers the pH was determined, and compared with the pi! obtained from the classic large-scale ISO extractions (Table 3). Tor most papers (here was a good agreement between the classic pi 1 extraction and the micro extraction. Again for the alkaline paper bad results were obtained. Also some acidic papers showed differences between both extraction methods. A possible explanation for this is the heterogeneity of the paper, since only a tew tens ot micrograms are sampled. On the other hand such heterogeneity may be deliberately shown by the micro pH measurement to reveal the diffusion of acidity for instance. Such ,111 opportunity was shown by measuring the pH as a function of the distance from the edge of a sheet ot paper (see Table 4 and Figure 4). At the same tune three strips of paper were sampled that covered the range of the microdrill sampling positions and their pH determined using the classic extraction method. Since the strips enclose several microdrills, it is expected that the pi I values ot the strips are situated in between the pH values of the microdrills taken at the edges of a certain strip. Results in Table 4 indicate that this is indeed the case. This does not only explain why in some cases divergent results are obtained using the microdrill technique, but also shows the high area specificity of the new miniaturized method. Moreover it is an indication ot the accuracy of the miniaturized pH determination method.
T able 5. pH-value and alkali reserve of some selectedcardboards used for the production of conservation boxes. Results are the average value of two measurements. The standard deviation is giren in brackets:
Figure 5. pH as a function of the alkali reserve for different conservation cardboards,