MOSSBAUER SPECTROMETRY APPLIED ON IRON GALL INK CORROSION
VERONIQUEROUCHON-QUILLET*1,CELINE REMAZEILLES2,
ALAIN WATTIAUX3, LEOPOLD FOURNES3
1LEMMA, Universite de la Rochelle,
avenue Michel Crepeau, 17042 La Rochelle Cedex,
Frame, Phone: (33) 5 46 45 82 17
e-mail:
2CRCDG, rue Geoffroy Saint Hilaire, 75005 Paris, France, Phone: (33) I 44 OS 69 90
3ICMCB, 87avenue du Doeteur Schweitzer,
33 608Pessac Cedex, France Phone: (33) 5 40 00 62 61
ABSTRACT
When used on paper, iron gall inks may provoke iron(II) catalysed oxidation and acid hydrolysis of the cellulose. In this work, some preliminary Mossbauer spectrometry measurements have been performed on laboratory probes leading to the conclusion that several kinds of iron(II) and iron(III) compounds are formed during the inking process. In particular, it has been observed that gallic acid has an influence on the iron(III)/iron(II) ratio, probably because of his reducing properties.
INTRODUCTION
The iron gall ink corrosion of paper is commonly attributed to cellulose oxidation mechanisms catalysed by iron(II) and to cellulose hydrolysis mechanisms induced by the high acidity of the ink. It has been noticed on original manuscripts that no direct correlation could be established between the iron content and the degradation state of the paper [1]. This point is noticeable even on inscriptions which have been written on the same sheet of paper, and thus preserved in a similar environment. One can argue that the acidic hydrolysis may differ from one ink to an other, as well as the iron(II) to iron(III) ratio. Also even if the iron sulphate used in the preparation of the ink has a drastic impact on the paper degradation, the influence of the other ink ingredients should not be neglected.
In this work, we explored the influence of the tannins, and especially gallic acid, on the oxidation state of iron in the ink. Gallic acid forms with iron(II) and iron (III) several chemical complexes which have already been largely studied [2]. But gallic acid, as many polyphenols, is also a powerful reducing agent. The redox potential of gallic acid/quinonic form (0.799V) is very close to that of iron(H)/iron(III) (0.749V). This means that gallic acid may undergo redox processes with iron(III), leading to the formation of iron(II) together with quinonic acid [3], Anti and pro-oxidative properties of gallic acid were already studied in solution systems [4], but no work has been done in this field on paper probes. For this reason, we undertake this work on laboratory paper probes, using Mossbauer spectrometry for the analysis of the iron environment.
EXPERIMENTALS
Preparation of the samples
All samples were prepared under laboratory conditions using the following pure products: gallic acid (ALDRICH n°39,822-5), and iron(II) sulphate heptahydrate (ALDRICH n°21,542-2). Two solutions were prepared: The first solution called "Fe" contains only iron sulphate with a concentration of 41.67 g/1. The second solution, called "Ac+Fe", was prepared with the same iron sulphate concentration and with a gallic acid concentration of 5.13 g/1. This latter solution presents an iron to gallic acid molecular ratio of 5.5. It contains also a great excess of iron, in regard to gallic acid. Some of this solution was dried at ambient atmosphere, and the resulting matter was considered as a reference sample called "dried ink".
The laboratory probes consist of sheets of paper immerged in the different solutions for approximately 10 seconds, then dried in an ambient atmosphere.
The selected paper is machine made from a combination of 95 % of cotton linters and of 5 %of softwood. It is supplied by the Institute of Industrial Technology, Nederland (TNO reference: PAPER 2 "Cotton linters cellulose paper"). This paper contains no size, no charge, no coating material, and is also very close to pure cellulose. Its average weight is close to 76 g/rrr
Some of these papers were artificially aged for 7 days in a Votsch 0020 oven with similar conditions to those used in previous work by Neevel [5]. Temperature remains constant at 90 °C, and the relative humidity changes every 3 hours, between 80 % and35 %.
Mössbauer spectrometry
Mössbauer measurements were performed at room temperature using a constant acceleration HALDER type spectrometer, with a 57Co source (Rh matrix) in transmission geometry. The spectra were recorded at 293 K. It was necessary to assemble 16 sheets of impregnated papers together to enable spectra recording in a reasonable timescale inferior to two weeks. Moreover, for such sample preparation linebroadening effects can be neglected. The velocity was calibrated using pure iron metal as reference material. The experimental data were resolved into symmetric doublets with Lorentzian lineshapes using an iterative least-squares fit program. When the refinement of the Mossbauer spectra showed an important and abnormal widening of the peaks, the spectra were fitted assuming a quadrupolar splittings distribution [6]. The best fitting of the experimental spectra leads to the determination of the following Mossbauer data : first the value of the isomer shift, called "δ" enables the determination of the oxidation state and the coordination of iron, then the quadrupolar splitting, called "Δ", gives an idea of the iron site distortion, and finally, the linewidth (Γ) gives a good indication on the order-disorder state of sample.
RESULTS AND DISCUSSION
The Fig. 1 and 2 report the Mossbauer spectra recorded at 293 K for all the probes. They characterize paramagnetic samples and each spectrum can be described by some quadrupolar doublets characteristic of iron(III) or iron(II).
The tables I and II summarize the Mossbauer data which were determined via the fitting procedure.
Looking to the isomer shift values, we can concluded that all the different iron sites correspond to octahedral environments. For further discussion, all these sites were labelled the following way: II-x or IH-x. The first part of the label (II or III) refers to the iron oxidation stale. The second part of the label (x) refers to an arbitrary letter and characterizes the site.
The dried ink
The ink was prepared with a great excess of iron sulphate in regard to gallic acid. The presence in the ink of a great quantity of iron sulphate (site Il-a) after the drying process was also quite expectable. No other iron(II) site was however detected, meaning that all the iron which participates to chemical reactions, is oxidized too.
The fit of the experimental spectra was not possible when assuming a single iron(III) site. Two iron(IIl) sites, showing very similar isomer shift values (Ill-a and Ill-b), were in fact necessary for a good adjustment ofthe calculated and experimental spectra. Some hypothesis relative to the identifications of these two sites will be formulated below.
Paper impregnated with the "Fe" solution
Two Fe(III) sites were detected. The first III-c site seems to be rather close to the III-a site or III-b site previously identified on dried ink. The similarity of these sites suggest that the Ill-a site may correspond to an iron(III) oxide, and not to the iron gall complex. This site appears to be quite stable since it is still detected in the same proportion after artificial ageing (III-c site).
The second II-d site shows quite different parameters from the previous reference sites. It is quite unstable and seems to evolve to the Ill-e site after artificial ageing. The high value of D determined on the III-e site accounts for a strong distortion ofthe iron(III) environment.
Paper impregnated with the "Ac+Fe" solution:
Here again two different iron(III) sites are identified. The first III-fsite is rather close to the III-b site. The similarity of these two sites suggests that the III-b site may correspond to the iron-gall complex. The second III-g site was not identified on previous references, and is very close to the III-e site. It seems moreover to be quite stable, since it is still detected after artificial ageing. Contrary to what is observed on paper impregnated with the "Fe" solution, a significant amount of iron(II) is detected. Its environment (II-b site) is quite different to that of iron sulphate (IIl-a site). The existence of divalent iron shows the important contribution of gallic acid to the redox iron(II)/iron(III) balance.
Fig. 1. Mossbauer spectra of probes immerged in"Fe" solutions.
(a) iron sulphate reference; (b) non aged paper probe; (c) paper probe after 7 days of artificial ageing.
Fig. 2. Mossbauer spectra of probes immerged in "Ac+Fe" solutions.
(a) dried ink reference; (b) non aged paper probe; (c) paper probe after 7 days of artificial ageing.
Tab. 1. Mossbauer data ofthe reference samples.
The impact of gallic acid
Table III reports the different proportions in iron(II) and iron(III) which can be drawn from table II data. It first shows that no iron(Il) was detected on paper probes impregnated with iron sulphate only. Even if the iron sulphate solution was freshly prepared and contained mainly iron(II), the iron deposited on the paper is fully oxidized into iron(III) during the drying stage.
After 7 days of artificial ageing, the probes impregnated with pure iron sulphate still contains less than 1 % of iron(II), meaning that there is no real reduction of iron(III) during the artificial ageing.
When gallic acid is present in solution, a significant quantitiy of iron(II) is detected in the probes after the drying process. This iron(II) can however not be identified to iron sulphate because the Mossbauer parameters ofthe II-b site differ significantly to those ofthe II-a site.
Tab. 2.Mosshauer data of the paper samples.
The redox reaction depicted on Fig. 3 could account for the presence of iron(II) in the probes: the redox potential of gallic acid is close to that of iron(II). and similar redox reactions happening in solutions have been pointed out in previous work [3|.
After 7 days of artificial ageing, the paper contains much less iron(II). since its ratio comes down to 9 %. During the artificial ageing, which hap-
Fig. 3. Redox reaction between gallic acid and iron sulphate.
Tab. 3. Iron(II) and iron(III) proportions in laboratory paper probes.
pens in the presence of oxygen, the iron(II) is oxidised into iron(III), and the reverse reducing reaction is probably limited by the degradation of gallic acid.
CONCLUSION
This preliminary work puts to evidence that the presence of gallic acid in the ink increases significantly the concentration of iron(II), thus enhancing the iron gall ink corrosion.
Mossbauer spectrometry proves also to be a powerful technique for the understanding of iron gall
ink corrosion mechanisms. Despite the low concentration of iron in the paper, some measurement could be undertaken in an efficient way.
One can expect, with this technique to draw much more information on the environment of iron and to identify the iron gall complex and the different iron oxides present in the ink. However, the first results presented here should be considered with great caution. More references are required to confirm the identification of the iron sites, and some further measurements will be undertaken at low temperature to improve the quality of the data.
REFERENCES
I Remazeilles, C. PhD. Thesis, University of La
Rochelle, July the 18lb (2001). 2. Krekel, C, in ,,Tintenfrassschaden und ihre
Behandlung", Fohlhammer, Stuttgart, 25-36
(1999).
3. Kipton. H., Powell, J., Taylor, M. C. Aust. J. Chem., 35, 739(1982).
4. Strlic, M., Radovic, T., Kolar, J., Pililar, B., J. Agric. Food Chem, 50, 6313 (2002).
5. Neevel, J. G., Reslaurator 16: 143-160 (1995).
6. Hesse, J., Rubartsch, A., J. Phys. E. Sci. Instrum, 7,526(1974).