Langmuir 2005, 21, 10743-10748

10743

Nanoparticles of Calcium Hydroxide for Wood Conservation. The Deacidification of the Vasa Warship Rodorico Giorgi, David Chelazzi, and Piero Baglioni* Department of Chemistry and CSGI, University of Florence, via della Lastruccia 3-Sesto Fiorentino, 50019 Florence, Italy Received March 12, 2005. In Final Form: August 24, 2005 The seventeenth century Swedish warship Vasa represents a unique case in the study of ancient wrecks and a challenge for finding new methods for artifacts conservation. The presence of sulfuric acid inside the wooden structure of Vasa is one of the possible causes of chemical damage of the wood. During recent investigations, pH values below 2 were observed inside the wreck in several places. Neutralization treatments temporarily raised the surface pH about 6 units, but after a few months the pH reverted back to original values. In this study we show that wood from the Vasa warship can be deacidified by using a dispersion in 2-propanol of calcium hydroxide nanoparticles. These particles can penetrate into the wood allowing a very efficient deacidification. Alkaline nanoparticles are converted into calcium sulfate without mechanical stress to the wood’s lumens. Additional applications produce an excess of alkaline nanoparticles that are converted into carbonate, an alkaline reservoir to protect the wood from further acid attack. Artificial aging of Vasa wood demonstrates that nanoparticles facilitate protection of wood toward further acid degradation.

Introduction The ship Vasa sank in its maiden voyage in 1628 and remained in the water of Stockholm harbor for 333 years. In 1961, it was recovered and treated for 17 years with poly(ethylene glycol) (PEG, H(OCH2CH2)nOH with 10 < n < 90) in order to prevent the wood from shrinking after drying, which would otherwise bring a dramatic loss in the physical and mechanical properties of the structure. The Vasa was the first major object for which aqueous solutions of PEG have been used. Intermittent spraying of glycol was followed by 9 years of slow drying before the ship was exhibited in 1990 in the Vasa Museum in Stockholm.1 PEG treatment is commonly used for many ancient shipwrecks that have been salvaged: examples are the Skuldelev Viking ships, the Bremen Cog, the Mary Rose (Portsmouth), and the Batavia (Western Australia).2 In the last few years, Vasa wood developed a consistent increase in acidity and conventional deacidification procedures failed. According to Vasa curators the development of acidity and, in particular, the production of sulfuric acid inside the wooden structure “threaten the continued preservation of the Vasa”.3,4 Sulfates originated from a large amount of elemental sulfur produced by the metabolic action of the bacteria living in the wasted anoxic water of Stockholm harbor. Oxidation of the reduced sulfur, coming from the penetration of hydrogen sulfide into the timbers, was shown to be catalyzed by iron species released from the completely corroded original iron bolts, as well as from those inserted after ship salvage, also subjected to a severe corrosion process.3 Indeed, these provided a new iron source in the PEG-treated wood. Oxidation of sulfur has been estimated to produce about 5 tons of sulfuric acid, and 2 tons were estimated to be already present in the wooden structure.3,4 * Corresponding author. E-mail: [email protected]. (1) Håfors, B. Conservation of the Swedish Warship Vasa from 1628. Vasa Stud. 2001, 18, 1-180. (2) Hoffmann, P.; Singh, A.; Kim, Y. S.; Wi, S. G.; Kim, I.; Schmitt, U. Holzforshung 2004, 58, 211-218. (3) Sandstro¨m, M.; Fors, Y.; Persson, I. The Vasa’s New Battle. Sulphur, Acid and Iron. Vasa Stud. 2003, 19, 1-80. (4) Sandstro¨m, M.; Jalilehvand, F.; Persson, I.; Gelius, U.; Frank, P.; Hall-Roth, I. Nature 2002, 415, 893-897.

The presence of iron represents one source of Vasa degradation. In fact, iron is involved in different reaction pathways, which seem to be strictly connected to each other. Besides increasing the reaction rate for oxidation of sulfur,5,6 iron catalyzes the oxidative degradation of cellulose-producing acid side chain groups (carboxyls).7 Iron is also probably active in the degradation of the end groups of poly(ethylene glycol), giving formic acid.8 Furthermore, iron ions interact with PEG,9 and it is wellknown that PEG is a solid-state electrolyte with efficient ion transport in the solid state.10 The overall result is an increase in acidity. Preservative treatments to arrest acid wood hydrolysis of the Vasa should focus on the removal of sulfur and iron compounds, which are the main causes of degradation. A recent neutralization treatment performed by Vasa curators with wet sodium carbonate/bicarbonate aqueous (washing soda/baking soda) poultices removed PEG and temporarily raised the surface pH from 1-3 to about 6-9. The neutralization is effective in the first 0.5 cm below the wood’s surface but leaves the bulk strongly acidic. Furthermore, alkaline treatment in aqueous solvent could be harmful to the wood fibers, since it delivers, at once, a large amount of hydroxyl ion, which may work through an alkali-catalyzed β-alkoxy elimination, leading to cellulose degradation.11-13 We propose here a new method for the neutralization of wood, based on the application of nanoparticles.14-16 A (5) Baird, C. Environmental Chemistry, 2nd ed.; W. H. Freeman: New York, 1999; pp 427-437. (6) Lowson, R. T. Chem. Rev. 1982, 82, 461-497. (7) Emery, J. A.; Schroder, H. A. Wood Sci. Technol. 1974, 8, 123137. (8) Kilmister, K. Preserving Our Past. An Investigation into Archaeological Wood from the Shipwreck of the Batavia. B.Sc. Thesis, Western Australia University, Perth, Australia, 2001. (9) Guilminot, E.; Dalard, F.; Degigny, C. Corros. Sci. 2002, 44, 21992208. (10) Gray, F. M. In Polymer Electrolytes; Connor J. A., Ed.; RSC Materials Monographs; The Royal Society of Chemistry: Cambridge, 1997; pp 1-136. (11) Kolar, J. Restaurator 1997, 18, 163-176. (12) Knill, C. J.; Kennedy, J. F. Carbohydr. Polym. 2003, 51, 281300. (13) Richards, G. N.; Sephton, H. H. J. Chem. Soc. 1957, 4, 44924499.

10.1021/la0506731 CCC: $30.25 © 2005 American Chemical Society Published on Web 10/06/2005

10744

Langmuir, Vol. 21, No. 23, 2005

Giorgi et al.

Table 1. pH Values of Fresh Wood and Samples of Wood from Vasa, Pre- and Post-Deacidification, Measured with a Glass Electrode sample

pH

sample

pH

sample

pH

sample

pH

Vasa pine untreated Vasa oak untreated

2.5

Vasa pine treated with soda solution Vasa oak treated with soda solution

9.0a

Vasa pine treated with nanoparticles Vasa oak treated with nanoparticles

6.7

fresh pine wood fresh oak wood

4.5-5.0

a

2.8

9.2a

4.5

4.0-4.5

Measurements performed over the surface with pH indicator paper.

similar method was very effective for the neutralization of paper and canvas acidity.17,18 Materials and Methods A. Samples. Oak and pine samples from the Vasa were received from the Swedish Maritime Museum authorities. For the characterization and treatment, wood was cut from the surface along the radial direction in cubes of 3 cm. Blocks of the same size were also cut from samples of fresh wood (oak and pine). The pH was measured by suspending 1 g of wood powder, collected by milling, in 5 g of water. The extracting suspension was not heated to prevent possible further acidification occurring from ester hydrolysis. The buffer capacity of the suspension increases with extraction time, due to the colonization of microorganisms or the conversion of acetylic groups to acetic acid. The pH does not change after 24 h. A glass electrode was used for the determination of the pH. This procedure is in good agreement with other methods when wood has good buffer capacity. In addition, the pH has been also locally determined with indicator paper (sensitivity of 0.5 units). B. Wood Deacidification. B.1. Treatment with the Soda Solution. Samples were treated following the same procedure as that of the Swedish curators of Vasa3. We applied poultices soaked in a baking soda/washing soda solution (NaHCO3/Na2CO3‚10H2O ) 7:3, 5% in water, pH ∼10) on the whole surface of the wood blocks. The alkaline treatment provided neutralization and removed PEG from the surface. The poultices were applied for 3 days. The pH of the wood blocks was 5.5 for oak and 8.0 for pine. The pH measurement, performed by using indicator paper on the blocks, gave slightly different results: a pH of about 9 was found in the treated surface (see also Table 1), and up to 0.5 cm from the surface. B.2. Synthesis and Application of Nanoparticles. Ca(OH)2 nanoparticles were obtained from homogeneous-phase reaction and used as a dispersion in 2-propanol.14,16,17 NaOH solution and CaCl2 solution were separately heated to the selected temperature in the range of 80-90 °C. When the selected temperature was reached, the two solutions were rapidly mixed under stirring keeping the temperature of the mixture constant within 1 °C. The supersaturation degree, defined as the ratio [Ca2+]/[Ca2+]sat, where [Ca2+]sat is the concentration of Ca2+ cations in the Ca(OH)2-saturated solution, was kept in the range of 2-10. The Ca(OH)2 suspension was allowed to gradually reach room temperature under a nitrogen atmosphere to avoid Ca(OH)2 carbonation. The supernatant solution was discarded, and the remaining suspension was washed 5 times with limewater to reduce the NaCl concentration to below 10-6 M. Each time, the dilution ratio between the concentrated suspension and washing solution was about 1:10. The complete removal of NaCl from the suspension was controlled by AgNO3 tests. The suspension was then concentrated in a vacuum at 40 °C up to a 0.8 Ca(OH)2/ water weight ratio. The wood samples (fresh and from Vasa) were washed for a week with distilled water, to remove most of the PEG, which would prevent particles from penetrating into the wood channels. PEG 600, 1500m, and 4000 are soluble in water and were (14) Ambrosi, M.; Dei, L.; Giorgi, R.; Neto, C.; Baglioni, P. Langmuir 2001, 17, 4251-4255. (15) Salvadori, B.; Dei, L. Langmuir 2001, 17, 2371-2374. (16) Bonini, M.; Bardi, U.; Berti, D.; Neto, C.; Baglioni, P. J. Phys. Chem. B 2002, 106, 6178-6183. (17) Giorgi, R.; Dei, L.; Ceccato, M.; Schettino, C. V.; Baglioni, P. Langmuir 2002, 18, 8198-8203.

extracted from wood. The washed blocks were soaked with a 0.05 M alkaline nanoparticle dispersion for 13 h. C. Physicochemical Characterization. The size and shape of the particles were determined by SEM, TEM (Philips XL20 and Philips EM201C, respectively), and AFM (not reported here), while the crystal structure of the obtained solid was investigated by X-ray diffractometry (XRD).14-18 The dimensions of the particles ranged from about 50 to 300 nm. XRD spectra were obtained on dried particles using a Philips PW1050/70 apparatus with a Co KR X-ray (λ ) 1.78 Å) source. The samples were prepared according to standard procedures. The DTGA curves of wood samples were recorded in dynamic, nonisothermal conditions, using a TA SDT Q600. The experimental conditions were as follows: temperature range, 30-500 °C; sample weight, 2-5 mg; heating rate, 10 °C/min; N2 flow, 100 mL/min. Wood blocks were sampled in areas up to 1.5 cm below the surface, along the radial direction, and powdered for the analysis. Cross-sectioned blocks from Vasa wood were investigated, after treatment with nanoparticles, by scanning electron microscopy (SEM), using a Philips 515 microscope. The samples were previously coated with a very thin gold layer deposited by sputtering under vacuum. EDS analysis was performed by using an energy-dispersive X-ray system, INCA x-sight, OxfordCambridge. Wood blocks from Vasa were powdered for XRD analysis after treatment with nanoparticles. The analysis was performed 3 h and 4 days after the application of nanoparticles, using an X-ray diffractometer Philips PW 1830 generator equipped with a Cu KR (λ ) 1.54 Å) source. About 1 mg of the dried wood powder was deposited, randomly oriented, onto a Plexiglass sample container, and the XRD patterns were recorded at a scan rate of 2°/min, under nitrogen atmosphere. D. Artificial Aging of Wood Samples. Acid aging of fresh wood samples (oak and pine) was carried out by keeping fine wood powder in a sulfuric acid solution (pH ) 3) for 1 h. Hydrothermal aging of Vasa wood blocks, washed with water and treated with alkaline nanoparticles, was carried out by storing the wood blocks under controlled temperature (80 °C) and 78% relative humidity (using a saturated KCl solution).

Results and Discussion The conservation of Vasa is a quite complex problem due to the possible chemical reactions that can develop or are already developed. We can summarize, in agreement with the Vasa curators’ investigations, that the most urgent problem to be solved is to slow the production of the acid inside the wood and possibly to remove iron (or make it a nonactive form) and sulfur. In the case of Vasa the situation is even more complicated by the presence of PEG and of the boric acid/borax (7:3) mixture used as fungicide in the PEG spray solutions applied to avoid the wood shrinking.4 Figure 1 reports a longitudinal cross section (from the surface of the wood to the interior) where both the PEG layer and the wood channels are clearly visible. The mapping of the elements (SEM/EDS analysis) of the cross section (see Figure 1) shows that iron is inhomogeneously distributed and concentrates at the wood-PEG interface, clearly indicating a preferential (18) Giorgi, R.; Dei, L.; Schettino, C.; Baglioni, P. In Works of Art on Paper, Books, Documents and Photographs: Techniques and Conservation; Daniels, V., Donnithorne, A., Smith, P., Eds.; IIC Press: London, 2002; p 69.

Nanoparticles for Wood Conservation

Langmuir, Vol. 21, No. 23, 2005 10745

Figure 1. SEM micrographs of a cross section of Vasa wood showing a thick layer of PEG covering wood fibers (A). Picture B reports the elemental distribution in the sample (red ) sulfur, green ) calcium, blue ) iron) obtained by SEM-EDS mapping. The distribution map for each different element is also reported (bar ) 100 µm).

iron-PEG9 interaction. This further complicates the conservation of the Vasa wood since it is well-known that PEG is a solid-state electrolyte with efficient ion transport in the solid state10 and iron is a catalyst for cellulose degradation.8,19,20 Our contribution to partly solve the conservation of Vasa is limited to the deacidification of wood. We believe that this contribution will be useful for both conservators and scientists working in the nanotechnology field. We report for the first time (as far as we know) a method to deacidify wood, based on alkaline nanoparticles. This method explores the possibility of using nanoparticles in nonaqueous solvents for wood conservation. Nanoparticles have been dispersed in 2-propanol, but the method is not restricted to this solvent, and fluorinated solvents can be used as well. A nonaqueous solvent is the key factor for the efficacy of the method. It prevents an instantaneous dissociation of the alkaline particles (allowing a good control of the release of the OH- groups in the wood) and provides a good penetration of the particles into the wood. Alcohols, with a surface tension that is up to 4 times lower than that of water, are able to wet the wood pores efficiently and quickly. Oak and pine wood samples from the Vasa warship have been deacidified by using two different neutralizing methods: (i) application of washing soda/baking soda solution poultices, as previously reported by Vasa curators;3,4 (ii) treatment with nanoparticles, after the removal of PEG with water. We also investigated oak and pine fresh wood and some artificially aged samples that have been compared to Vasa wood. The acidity of fresh wood is due to the presence of free or bound organic acid species in the extractives, hemi(19) Emery, J. A.; Schroeder, H. A. Wood Sci. Technol. 1974, 8, 123137. (20) Johansson, A. Air Pollution and Paper Deterioration. Ph.D. Thesis, Go¨teborg University, Goteborg, Sweeden, 2000.

cellulose, and polyphenols: extractives contribute 80% of the total acidity in oak wood.21,22 The pH of fresh wood varies from 4 to 7, depending on the wood species; oak is one of the most acidic. The application of calcium hydroxide nanoparticles to Vasa wood increased the pH of the samples. Table 1 reports a ∆pH increase of 1.7 units for oak and 4.2 units for pine with respect to the untreated Vasa samples. The difference in the pH change is related to the greater intrinsic acidity of oak wood with respect to that of pine. The final values, reached after nanoparticles treatment, are very close to that of the fresh oak (4.5) and about 2 units greater for pine (about 6-7). Vasa samples treated by using soda exhibited very high surface pH (around 9). However, this treatment did not change the pH below the wood surface, and just 0.5 cm below the wood surface the pH of the sample remained acid (from pH ) 2-4, depending on the wood species and the sampling). Probably, the exposure of samples to a very alkaline aqueous solution produces a consistent superficial deacidification effect, but many hydroxyl groups could be also consumed in different reaction pathways, some of them involving the remaining cellulose of the Vasa wood. Penetration of particles through the wood pore structure was followed by means of SEM analysis. The pictures were taken on a cross-sectioned sample 1.5 cm below the surface, along the radial direction. Figure 2A shows the fiber walls of wood after the treatment with nanoparticles. Figure 2B shows the presence of calcium sulfate particles, probably produced by the interaction of the hydroxide nanoparticles and the sulfuric acid, present in the wood. Nanoparticles are orders of magnitude smaller than the wood lumen (capillary tubes for water transport), and even an increase in their volume, such as that due to transformation of the hydroxide into sulfate, cannot generate any mechanical stress. EDS chemical analysis (not (21) Balaban, M.; Ucar, G. Holz RohsWerkst. 2001, 59, 67-70. (22) Gindl, M.; Tschegg, S. Langmuir 2002, 18, 3209-3212.

10746

Langmuir, Vol. 21, No. 23, 2005

Giorgi et al.

Figure 3. XRD patterns of Vasa wood, recorded 3 h and 4 days after the deacidification treatment with calcium hydroxide nanoparticles.

Figure 2. Scanning electron microscopy (SEM) pictures: (A) Ca(OH)2 nanoparticles on the wall fibers of wood, 1.5 cm below the wood surface (bar ) 20 µm); (B) calcium sulfate formed after deacidification (bar ) 20 µm); (C) calcium carbonate alkaline reservoir (bar ) 6 µm).

reported) evidenced the presence of calcium (presumably indicating the presence of carbonate or hydroxide) in the platelet particles (see Figure 2A-C) and calcium combined with sulfur in the acicular crystals shown in Figure 2, parts B and C. However, EDS is not suitable to unambiguously identify the type of the nanoparticles containing calcium alone. XRD on deacidified and dried oak wood samples (Figure 3), performed 3 h and 4 days after the application of nanoparticles, showed the presence of calcium hydroxide (as portlandite) and its complete conversion to calcium carbonate (as calcite). This reaction was also evidenced by FT-IR analysis performed with the KBr pellet technique (results not reported here). A further characterization of the efficacy of the deacidification process has been performed by differential thermogravimetric analysis, DTGA. DTGA is a useful and simple method for characterizing heterogeneous organic materials and can be conveniently used to check the degradation status of cellulose molecules before and after the deacidification treatments. According to the literature, the thermal degradation of the macromolecular structure of wood components occurs in the 160-390 °C temperature range. High temperatures for thermal reactions are associated with a better degree of conservation of wood and cellulose.23-25 Table 2 sum-

marizes the DTGA data for Vasa wood (oak and pine, before and after the deacidification treatment) and for the fresh and the artificially degraded fresh wood. In the fresh oak wood, pyrolysis of cellulose occurs at 342 °C. In PEG-treated Vasa oak wood the peak is shifted to a lower temperature (325 °C) with a ∆T of 17 °C. For the Vasa pine wood the ∆T is 28 °C. Fresh wood artificially aged using sulfuric acid (see Materials and Methods, section D) shows a ∆T of about 30 °C. The temperature shifts can be explained by considering that the acidification and the oxidation depolymerize the cellulose chains producing a large fraction of lower molecular weight oligomers; therefore, less energy is required for the thermal decomposition. In many cellulose-based materials, such as paper and cotton, shifts of about 30 °C are characteristic of very degraded samples, where the degree of polymerization of cellulose is 15% of the standard value.24 Figure 4 shows the DTGA curves of Vasa wood (oak and pine) after treatments with the two methods (washed with baking soda solution (I), and treated with alkaline nanoparticles after washing with distilled water (II)). Washing with water and soda treatment on oak wood seems to provide no improvement in the thermal resistance toward degradation of the samples. The maximum pyrolysis temperature is 323 °C in the washed oak curve and 326 °C in the soda-treated one, very close to the values of the untreated samples. The curve of the sample treated with soda shows a peak at 365 °C due to the pyrolysis of very degraded PEG. PEG 1500, in normal conditions and in nitrogen atmosphere, shows a maximum pyrolysis temperature at 413 °C. The treatment with nanoparticles reverts the treated wood to a thermal behavior very similar to that of fresh oak wood. In fact, the maximum pyrolysis temperature is 344 °C, the same as fresh oak. This does not mean that cellulose has reverted to its original status, but after the nanoparticles treatment wood seems “more resistant” to the thermal degradation. We hypothesized that the filling of the pore structure of wood with nanoparticles creates safer conditions in the immediate surroundings of the cellulose molecules and the cellulose hydrolysis is disfavored, preserving the remaining cellulose of the Vasa wood (23) Sandu, I.; Brebu, M.; Luca, C.; Sandu, I.; Vasile, C. Polym. Degrad. Stab. 2003, 80, 83-91. (24) Vicini, S.; Princi, E.; Luciano, G.; Franceschi, E.; Pedemonte, E.; Oldak, D.; Kaczmaerk, H.; Sionkowska, A. Thermochim. Acta 2004, 418, 123-130. (25) Soares, S.; Camino, G.; Levchik, S. Polym. Degrad. Stab. 1995, 49, 275-283.

Nanoparticles for Wood Conservation

Langmuir, Vol. 21, No. 23, 2005 10747

Table 2. Pyrolysis Temperature of Cellulose, for Oak and Pine Woods from Vasa and for Fresh and Artificially Acidified Fresh Wood, Measured by Differential Thermogravimetric Analysis (DTGA)a samples

fresh wood

fresh acidified wood

untreated Vasa wood

Vasa wood washed with water

Vasa wood treated with soda solution

Vasa wood treated with nanoparticles

oak Tmax (°C) pine Tmax (°C)

342 363

310 334

325 335

323 355

326 295

344 360

a

Error ( 1 °C.

Figure 5. Plot of the pyrolysis temperature of cellulose vs the artificial hydrothermal aging time for oak Vasa wood washed with water (triangles) and treated with Ca(OH)2 nanoparticles (full circles). Lines are guides for the eyes.

Figure 4. Differential thermogravimetric analysis (DTGA) of oak (A) and pine (B) wood from Vasa, treated with different methods, and of fresh wood.

from thermal degradation. Figure 4 also reports the DTGA for fresh oak wood. The pyrolysis of hemicellulose in the hemicellulose-lignin matrix, presents a shoulder at 290 °C in the cellulose pyrolysis peak. This peak is missing in the Vasa wood suggesting that most of the hemicellulose is already degraded in the waterlogged oak wood.26 For Vasa pine wood, the washing treatment and the partial removal of PEG partly improved the thermal resistance. The DTGA curve shows a pyrolysis temperature of 355 °C, followed by the peak of the residual PEG pyrolysis at 380 °C, indicating a lighter PEG degradation with respect to the previously mentioned samples. In the pine sample treated with soda, the peak from PEG is present at about 380 °C, and cellulose has a very broad and flat band with a maximum at about 295 °C, indicating a consistent degradation. Similarly to oak wood, the soda treatment is not effective in restoring the thermal behavior of the pine wood. On the other hand, the treatment with alkaline nanoparticles is effective in restoring the original thermal (26) Bettazzi, F.; Giachi, G.; Staccioli, G.; Chimichi, S. Holzforschung 2003, 57, 373-376.

behavior of pine wood. The curve of the sample treated with nanoparticles shows the maximum pyrolysis temperature at 360 °C, very close to that of the fresh pine value (363 °C). As for the oak sample, the peak related to hemicellulose pyrolysis was not observed simply because most of the hemicellulose was lost during the 333 years on the sea bed due to microbial activity. To test the efficacy of the deacidification treatment, Vasa wood blocks treated with calcium hydroxide nanoparticles and simply washed with water were hydrothermally aged (see Materials and Methods). This is a wellaccepted method to test the efficacy of deacidification protocols on cellulose-based materials.27,28 We monitored the maximum pyrolysis temperature of the blocks as a function of the aging time. The trend of the pyrolysis temperature is reported as a function of time in Figure 5. Figure 5 shows that wood treated with nanoaprticles presents after 120 h of thermal aging a pyrolysis temperature 10 °C higher than that of the washed wood. The improvement in the thermal behavior of the treated wood is evident, indicating that the positive effects of the treatment on the thermal resistance of wood last with time. It is worth recalling that the aging conditions used (80 °C and 78% relative humidity) are very severe. Even though a precise conversion of the artificial aging to the real aging time is difficult, it seems plausible that nanoparticle treatment facilitates a long-term protection of wood toward degradation. Conclusions This study reports the application of nanoparticles of calcium hydroxide to wood for deacidification. Nanoparticles are used from a nonaqueous solvent (2-propanol). Oak and pine wood samples from Vasa have been deacidified with nanoparticles and compared to the traditional washing soda/backing soda method. We found (27) Bansa, H.; Hofer H. H. Restaurator 1984, 6, 21-60. (28) Porck, H. J. Rate of Paper DegradationsThe Predictive Value of Artificial Aging Tests; European Commission on Preservation and Access: Amsterdam, 2000.

10748

Langmuir, Vol. 21, No. 23, 2005

that alkaline earth hydroxide nanoparticles (mainly calcium hydroxide, but this method is not specifically restricted to this metal), penetrating into the wooden structure, adhere to the wall fibers after the volatilization of the dispersing medium (propanol) and leave an alkaline reservoir. The slow release of basic groups from the hydroxides provides a harmless deacidification since the hydroxides once released quickly neutralize the acid, and the excess (if present) is converted into carbonates that are neutralizing agents as well. Deacidification with alkaline nanoparticles of Vasa wood improves important parameters related to the degradation status of cellulose molecules for both oak and pine samples. In fact, nanoparticles improve the thermal behavior of the degraded wood and the pH is reverted to values close to those of fresh wood. Artificial aging of Vasa wood shows that nanoparticles increase the resistance of wood to further degradation. Soda poultices remove most of the degraded PEG from the surface and partly neutralize the wood, but the treatment is not durable, and the thermal behavior of the treated wood is not improved. Probably, this depends

Giorgi et al.

on the lack of an alkaline reservoir (calcium carbonate) on the wood. This reservoir facilitates a long-term protection for the wood, as clearly showed by the artificial aging. Although the nanoparticle deacidification method should be scaled up to verify its applicability to Vasa conservation, we believe that it is a good starting point for a “real prevention” of Vasa wood from further degradation. Acknowledgment. We thank Ingrid Hall-Roth and Emma Hocker (from Vasa Museum) for providing wood samples from Vasa and Lars Ivar Elding (University of Lund) for helpful and stimulating discussions. We also thank Professor Mats Almgren (University of Uppsala), Luigi Dei, Alessio Becheri, Marco Lagi, and Ilaria Santoni (University of Florence) for useful discussions. CSGI and the Ministry for Education and Research (MIUR, PRIN2003) are also gratefully acknowledged for partial financial support. LA0506731