Analysis of historical papers
Although opacity values are very high and range from 90 to 98% (Fig. 9b), it has been shown that papers whose thicknesses (Fig. 4b) are lower also have lower opacity, as in examples 4D, 6D, and 8D. Such a conclusion is very logical because a smaller amount of fibers and fillers in the paper certainly contributes to greater transparency and lower opacity. Comparing the values of brightness and yellowness showed that they are inversely proportional, which means that when the brightness value is higher, the value of yellowness is lower and vice versa.
The brightness in the paper could depend on the proportion of filler that fills the space between the fibers in the paper as shown by a comparing the results of the XRF analysis with the brightness value. Thus, this comparison shows that samples with a higher proportion of calcium (Table 6) also have a higher brightness (Fig. 7b). This is best recognized in samples 1D, 5D, 6D, and 8D. Such a result suggests the opposite situation as well, i.e. that a lower brightness indicates a lower amount of calcium as the basic element of the filler, which is visible in samples 2D, 3D, 4D, and 9D. The exception is the 10D sample, which is visibly dark and has a brightness value of 18.16% and a yellowness value of 60.55% (Fig. 8b), while the calcium content according to the XRF analysis is high (Table 6). The cause of such values may lie in the aging of the paper or poor conditions in which the sample was stored, and not in the calcium content. An exception is also the 7D sample, whose brightness value is 65.32%, while its calcium content is low, which indicates that other factors also affect the brightness. In this case, it is possible that the better storage conditions are the reason why the paper has remained lighter.
The pH of the paper can also be related to the calcium content in the paper or filler that creates an alkaline environment. Thus, samples 2D, 3D, and 9D have an acidic pH between 5 and 5.45 (Fig. 6b), and the very low calcium content seen from XRF spectrum (Table 6). In the case where the calcium content is higher, the pH is neutral, around 7 [5]. This can be seen in samples 1D, 5D, 6D, and 8D. The gloss of paper surface (Fig. 10b) can be related to lower brightness and acidic pH, as obtained in samples 2D and 3D, where the measured gloss values were less than 1.
Spot tests performed for identification of starch in historic papers did not yield a positive result, proving that starch was very rarely used as a surface sizing additive from the 16th − 19th centuries on papers produced in Europe [8 ].
A test to identify lignin with the reagent phloroglucinol indicated the presence of fibers that have a slightly higher proportion of lignin. Samples 2D and 6D had the most colored fibers, which could indicate the presence of lignin, while in the sample 10D from the 16th century there are no such fibers present. A paper absorbency test conducted with a drop of water showed that the most absorbing paper samples were 1D and 2D, and the least 5D and 6D. The results of the ATR-FTIR analysis showed that it is the 1D and 2D paper samples that have no additives of either gelatin as a coating or calcium as filler, which could be the reason for their higher absorbency.
Microscopic Analysis was performed on the surface of the paper samples and on the fibers. Analysis of the surface of the paper revealed specific fibers that stood out, so it could be concluded that most of the specific fibers are thick light brown fibers that look like straw [23, 27]. Also, a thin blue fiber was found in samples 1D, 2D, 5D, 6D, 8D, and 10D. The presence of blue dyed fibers was confirmed by analysis on an optical microscope in samples 1D, 2D, 3D, 4D, 5D, 6D, and 10D. Based on morphology, the blue fibers in samples 1D, 2D, 3D, and 5D are cotton fibers while in samples 4D, 6D, and 10D they are silk fibers. According to the data from the literature, it is assumed that the blue fibers were dyed with indigo dye, which, in the period from the 16th to the 19th century, was used in Europe for dyeing fabrics [32]. Most of the fibers in all samples were recognized as flax and hemp, the most represented fibers cultivated for the production of textiles in Europe in the period from the 16th to the 18th century, while cotton was less represented. Cotton was identified in only two samples from the 19th century (5D and 6D). Such a result can be supported by the historical fact that, in the 16th-17th centuries, cotton was mostly imported from India and then hand-spun in Europe [23, 11].
ATR-FTIR spectra confirmed the presence of both types of fillers in most samples except in 1D and 2D, which is significant, but when considering the purpose of these papers, such results are not surprising. 1D is the paper Valvasor used as the base for his collection of prints while 2D served as the blotting paper for the ink that was placed under the paper used for writing. Calcium as an element is visible in both elemental analyses of inorganic elements (Tables 5 and 6) except in SEM-EDS in three samples: 3D, 4D, and 9D. Comparing the results with XRF, it is clear that the calcium peaks on these samples are smaller, which would indicate that the calcium content is lower and which can be further related to other analysis results such as pH, absorption, brightness, etc.
In the SEM-EDS and XRF analyses, many trace elements have been recorded that can be related to the composition of the paper (Tables <link rid="tb5">5</link> and 5). Thus, sulfur (S), potassium (K), and in some samples iron (Fe) are visible in XRF spectra, while aluminium (Al) and copper (Cu) are visible in SEM-EDS spectra. Sulfur is an element that is part of the molecule of gypsum (CaSO4 x H2O) and alum (KAl(SO4)2 x 12H2O) as well as aluminium and potassium. Iron as an element is visible in traces in all XRF spectra and can be attributed to the composition of iron alum (FeAl(SO4)2 x 12H2O). Silicon (Si) is an element that was detected in both elemental analyses and, according to the literature, is associated with the presence of straw. Straw stalks contain silicon oxide (SiO2) [25], 4–7% in wheat straw [21]. The presence of most of the elements recorded in traces during elemental analyses could be interpreted, though for some, such as manganese (Mn), magnesium (Mg), titanium (Ti), copper (Cu) and nickel (Ni), we did not find any explanation for their presence. Such a result could be attributed to sample contamination.
Analysis of paper sheets from Valvasor’s collection
With a comparison of the results of analyses performed on the paper sheets in VC and HP some similarities, as well as differences were observed. The thickness measurements revealed (Fig. 4) that the values range from 0.16 mm to a little more than 0.24 mm. The exception is the sample 5ND (collection of geographical maps [33]) where individual papers are glued to the fabric, so the measurements were carried out together with the fabric, and, as such, the result can be neglected. Although, afer we analysed only a small amount of historical papers, some conclusions could be made. It could be concluded that the thicknesses of the paper samples from the 16th and 18th centuries (10D, 5D, 6D) were somewhat lower compared to the papers from the 16th − 17th centuries (10D, 3D, 9D, 1D). Papers from the 19th century (2D, 7D, 8D) can have different thicknesses as a result of the development of production technology and the possibility of making different types of paper. Barrett in his research [5] shows that the thicknesses range from 0.15 mm to 0.3 mm and that towards the 19th and 20th centuries, thickness values increased.
The highest difference in properties between two set of samples was obtained for acidity/alkinity of papers (Fig. 6), probably due to different approaches to measurement. Measurement of paper pH on the surface with a drop of water, according to the TAPPI standard T529 [14], performed on HP fragments is compared with a non-destructive pH measurement directly on the surface, without a drop of water. It should be emphasized that later results, performed on the paper sheets from VC should not be taken as such, but should be considered only as an indicative value. In the non-destructive approach, the obtained pH values were in the acidic range, while in the destructive approach these values were broader. A comparison with the results of elementary analyses of SEM-EDS and XRF on paper fragments shows that the pH depends on the proportion of calcium or fillers in the paper. In this research, two things can be concluded - firstly, that the difference in results arose from different approaches to measurement and, secondly, that the papers from the 17th century have a lower proportion of calcium or fillers in the paper. There is another theory put forward by William Barrow [4] which was confirmed by Irene Brückle [34], where they stated that papers in the 17th century were coated with gelatin and alum, which remained on the surface of the paper due to which the surface of the paper has an acidic pH.
Observing the results of the brightness and yellowness (Figs. 7 and 8), a scattering of results may be noticed, the cause of which are different papers in the book marked 5ND, although most of the measured brightness values range from 40–60%. Comparing these values with the values of brightness of the tested samples historical papers fragments, it can be seen that the values in the 16th century were lower, then increased during the 17th century, and dropped a little during the 18th century, and, finally, increased in the 19th century. An attempt was made to explain the reason for such brightness in the proportion of calcium that could affect the brightness or yellowness of the paper. If we look at the oldest tested sample, i.e. paper from the 16th century (10D), it can be seen that the brightness is very low (18.16%) (Fig. 7), while the calcium content is high (Tables 5 and 6). This indicates that we cannot fully confirm the claim that a high brightness also means a high proportion of calcium or fillers in the paper. The reason for the low brightness values despite the high calcium content should be sought in other factors that influence the brightness. The yellowness (Fig. 8) is inversely proportional to the brightness, so the scatter of the results at paper sheets in VC is repeated, as well as the yellowness in HP samples being high and then dropping in the in 17th century, after which it slightly increases and drops again in the 19th century.
The values of opacity determined in the paper sheets in the VC (Fig. 9) are very similar, around 90%. The situation is similar with the opacity of HP fragments shown through the centuries where values range from 90–98%. The comparison of the opacity and thickness over the centuries showed that they coincide; thinner papers have lower opacity and vice versa.
The gloss of paper sheets in VC (Fig. 10) is low, in the range between 1 and 1.5. Again, values determined for paper sheets in book marked 5ND (collections of geographical maps) show more scattering, on some papers the values are up to 2.5. Observing the diagram of HP fragments through the centuries, a slight increase in gloss towards the 19th century was noted, with the exception of blotting paper marked 2D. The low gloss is connected with a paper surface which is not coated or sized.