Tables and Figures have independent numbering and are shown in separate groups at the end.
E.g. Fig.1, Fig. 2 , Fig. 3 ..
Table 1, Table 2, Table 3 ….
I miss the story of the ship.
Degradation of Marine Archaeological Wood Through Acidification
Sara Goldman
Office of Science, Science Undergraduate Laboratory Internship (SULI)
University of Colorado at Colorado Springs
Stanford Synchrotron Radiation Lightsource
Stanford, CA
August 20, 2010
Prepared in partial fulfillment of the requirements of the Office of Science, Department of
Energy's Science Undergraduate Laboratory Internship under the direction of Dr. Apurva Mehta,
Stanford Synchrotron Radiation Lightsource.
Participant:
_______________________________
Signature
Research Advisor:
_______________________________
Signature
TABLE OF CONTENTS
Abstract ……………………………………………………………………………………… ii
1.0 Introduction ………………………………………………………………………………….
2.0 Materials and Methods ……………………………………………………………………….
3.0 Results ………………………………………………………………………………………..
4.0 Discussion ……………………………………………………………………………………
5.0 Conclusion ……………………………………………………………………………………
6.0 Acknowledgements …………………………………………………………………………..
References ……………………………………………………………………………………
Figures ……………………………………………………………………………………….
ABSTRACT
Degradation of Marine Archaeological Wood Through Acidification. SARA M. GOLDMAN
(University of Colorado at Colorado Springs, CO 80919) APURVA MEHTA (Stanford
Synchrotron Radiation Lightsource, Menlo Park, CA 94025
The presence of sulfur in marine archaeological wood presents a challenge to
conservation. Upon exposure to oxygen in the air, sulfur compounds in waterlogged wooden
artifacts are being oxidized, producing sulfuric acid. This speeds the degradation of the wood,
potentially damaging specimens beyond repair. X-ray absorption spectroscopy was used to
identify the species of sulfur present in samples from the timbers of the Mary Rose, a preserved
16th century ship known to experience this acidification. Results showed sulfur content varied
significantly on a local scale, and identified correlations between the presence of inorganic
sulfurs and sulfates. Only certain species have the potential produce sulfuric acid by contact
with oxygen and seawater in situ, such as iron sulfides and elemental sulfur. Organic sulfurs,
such as the amino acids cystine and cysteine, are integral parts of the wood’s structure and
should not be removed. Due to the varying sulfur content in waterlogged wood, contact treatment
strategies for removing inorganic sulfur or neutralizing acid should pursue a more
comprehensive study to ascertain the amount and location before proceeding.
1.0 INTRODUCTION
Waterlogged archaeological wood, frequently in the form of shipwrecks, is being
excavated for historical purposes in many countries around the world. Even after extensive
preservation chemicals and techniques have been applied, scientists are discovering the
accumulation sulfate salts and measuring acidic pH on the surfaces of their artifacts. Sulfuric
acid degrades structural fibers in the wood by acid hydrolysis of cellulose, accelerating the
decomposition of the ship timbers [1]. Determining the sulfur content of waterlogged wood is
now of great importance in maritime archaeology. Artifact preservation is often more time
consuming and expensive than the original excavation; but it is key to the availability of objects
for future study as well as maintaining the integrity of historical data and the preserving the value
of museum pieces [2].
Sulfur occurs in a wide number of oxidation states from -2 to +6, and appears in
numerous organic and inorganic compounds in nature. However, it is a very minor component
of wood. X-ray absorption spectroscopy (XAS) is necessary due to its ability to detect a small
amount of a specific element among greater quantities of others. The sensitivity of XAS can also
distinguish between different oxidation states and identify ordered molecules. These minutiae
are not always recognized by other chemical means.
2.0 MATERIALS AND METHODS
2.1 The Sample
The sample of waterlogged archaeological wood was from the Mary Rose, currently in
Portsmouth, England. The piece was cut from the oak stem post, which was excavated in 2003,
and had not been treated with any preservation chemicals or any conservation methods other than
drying. The sample was approximately 3cm by 4cm by ½ cm.
2.2 Data Collection
Sulfur K-edge X-ray absorption spectra were collected on a 20-pole 2.0 Tesla wiggler
unfocussed beamline 4-3 at the Stanford Synchrotron Radiation Lightsource (SSRL) under
standard ring operating condition of 3 GeV and ~200 mA ring current. X-ray energy was varied
using a Si(111) double-crystal monochrometer with a non-fixed exit slit. Other upstream optics
included a Ni-coated vertically collimating and harmonic rejection mirror. The sample chamber
contained a helium atmosphere. The sample was stored and handled at room temperature. The
emitted X-ray fluorescence was measured at 0° using a Lytle detector. Thin double-stick sulfurfree Mylar tape was used to affix the sample piece to the sample holder. Energy calibration was
achieved using Na2S2O3·5H2O as the calibrant, which was run before and after sample scans.
The first peak of the Na2S2O3·5H2O spectrum was fixed at 2472.02 eV for calibration purposes
[3].
Additional S K-edge standard data were previously collected on beamline 4-3 by Dr.
Rimukta Sarangi.
2.3 Data Analysis
The data were individually reviewed visually in Sixpack, and any faulty or incomplete
scans removed. The scans were calibrated and averaged in EXAFSPAK. A second-order
polynomial was fit to the pre-edge region and subtracted from the entire spectra as background.
The data were normalized using the Pyspline program by subtracting a cubic spline and
assigning the edge jump to 1.0. The area under the pre-edge peaks was fit using IFEFFIT in
Sixpack, as well as the EDG_FIT subroutine in EXAFSPAK. The edge jump of the fifteen
positions, corresponding to the total amount of sulfur, was calculated with Sixpack.
Due to the thickness of the sample (±1/2 cm), the resulting spectra were observed to be
self-absorbed. Self-absorption correction was achieved using the FLOU program assuming
infinite thickness.
3.0 RESULTS
Fifteen spectra were taken at different locations on the sample piece what was the cross
section of the beam?. Although the spectra varied in total sulfur amount and species, there were
two prominent peaks occurring in all of them. FIGURE 1 Figure The first peak occurred at
2472.96eV. The second larger peak occurred at 2482.60eV.
Data were fitted with the sulfur standards what is ‘sulfur standards”. The thiol and
methionine standards produced results so similar that they could be used interchangeably.
Methionine was selected to act individually rather than including both standards. A majority of
the samples were fit with five components: methionine, elemental sulfur, sulfoxide, sulfonate,
and sulfate. Presumably these are sulfur compounds Total sum and R-value (the misfit scaled to
the data itself) were recorded with the fit fraction of the species present as shown in Table 2.
TABLE 2 For the sample with the least total sulfur content, best fits were achieved with
fractions of 0.1836 methionine, 0.4201elemental sulfur, 0.1139 sulfoxide, 0.0387sulfonate, and
0.1847 sulfate. FIGURE 3 For the sample with the greatest total sulfur content, best fits were
achieved with 0.0249 methionine, 0.0832 elemental sulfur, 0.0211 sulfoxide, 0.0608 sulfonate,
and 0.7150 sulfate. And the rest? Uncertainties? Do year know the number to 3 or 4 decimal
places?
The fit data were normalized by multiplying the edge jump (total sulfur) by the fit
fraction to get the species amount proportional to the total amount of sulfur present. The first
peak at 2472.96eV is composed of elemental sulfur and methionine. The second peak at
2482.60eV is composed primarily sulfate and to a lesser extent sulfonate. TABLE 4
4.0 DISCUSSION
The edge jumps calculated in the XAS spectra fluctuate from a low of 0.1915 to a high of
0.6897, indicating that the total sulfur content varies by up to three times on a local level.
4.1 Methionine/thiol What is this stuff
Of the fifteen samples, the point containing the greatest amount of methionine reached
2.7 times the location with the lowest content. The intensity of the XAS features is sensitive to
the concentration of the sulfur species [4]. The peak intensity for this compound would depend
on the physical amount of cell matter penetrated by the X-ray at the sample location, which
differed due to the irregular thickness of the sample, the cell structure of the wood, and the
underwater exposure it received. Regardless of the amount variation in different locations, the
largest quantity made up less than one fifth of the total sulfur content in that position. The sulfur
in the amino acids are a minor component and do not fluctuate greatly based on the total sulfur
content. Do the fractions in the table depend on measuring problems, e.g. absorption due to
thickness of samples
4.2 Sulfoxide What is this stuff
Sulfoxides occurred in a small but noticeable amount in all of the samples, and appear as
the minor feature between the first and second peak on figure 1. Sulfoxides are oxidized sulfides,
but sulfides to not naturally occur in oak. This suggests that the sulfoxides are an indicator of an
external source of sulfide. Whether oxidation occurred by contact with dissolved oxygen while
under water or upon exposure to the air after excavation is unknown. Sulfoxide does not vary
substantially based on the total sulfur content.
4.3 Sulfonate
Sulfonate is the salt of sulfonic acid, which is similar to sulfuric acid in structure.
Sulfonic acid can be produced through the slow oxidation of thiols (or methionines) with contact
to air and water. The surviving timbers of the Mary Rose are known to have been embedded in
clay and buried with silt, resting in an anaerobic environment in situ [8]. However, due to the
430 year burial process, some surfaces certainly came into contact with more dissolved oxygen
than others. Our component analysis shows that thirteen of the fifteen samples contain small
amounts of sulfonate, and the amount does not change greatly based on the total sulfur content.
4.4 Elemental Sulfur
Elemental sulfur was present in all fifteen samples as the primary component of the first
peak observed at 2472.96eV. Elemental sulfur is capable of reacting with air and water to form
sulfuric acid. Its current presence in the sample indicates that it was lacking sufficient oxygen
while under water and lacking adequate water after excavation to completely acidify. Elemental
sulfur does not occur naturally in wood or seawater, suggesting it came from an external source.
It tended to decrease slightly based on the total sulfur content.
4.5 Sulfate
Sulfate is the salt of sulfuric acid, and it appeared as the primary component of the total
sulfur content in all of the samples. In fact, as total sulfur content increased, the amount of
sulfate was the principal source FIGURE 5 . Trees get the sulfur they require from the soil they
inhabit by taking up inorganic sulfate, reducing it, and incorporating it into amino acids such as
methionine [5]. Wood does not naturally contain sulfate, all of it had to come from an external
source.
The features on the XAS spectra beyond the second peak at 2482.60eV indicate the
presence of an unidentified ordered sulfate. FIGURE 1 The more level features beyond the
second peak indicate a disordered “free” sulfate. A large portion of the free sulfates have bound
to cations to form ordered sulfates .
4.6 Bacterial Introduction
Three of the five possible sulfur components found in the sample come from outside
sources. Previous studies have identified the activity of anaerobic bacteria in the wood [6]. The
bacterial breakdown of organic matter produces hydrogen sulfide (H2S). H2S reacts with the
wood and its surroundings in a number of ways. When combined with seawater and dissolved
oxygen, it can form sulfuric acid immediately. Alternately, if there are free sulfates already
present in the water, it H2S can react with it to form elemental sulfur, which can then itself form
sulfuric acid when combined with dissolved oxygen and seawater. In addition, the ship
contained thousands of iron nails and fixtures, which over time introduced iron ions into the
surrounding water. Hydrogen sulfide reacting with these iron ions produces iron sulfides such as
pyrite and pyrrhotite. These iron sulfides can either produce sulfuric acid upon contact with air
and seawater, or elemental sulfur and iron precipitates if acid is already present [7].
1.0 Iron
Previous studies have identified iron sulfides as a potential sulfur component in the wood of the
Mary Rose [6]. Our sample spectra did not show any iron sulfides. However, it was observed in
the laboratory that when pyrite and pyrrhotite powder were mixed with seawater and exposed to
oxygen in the air, an acidic pH was measured within minutes for both.
5.0 CONCLUSIONS
All of the sulfur species capable of becoming sulfuric acid upon contact with air and
seawater were introduced as a result of the activities of anaerobic bacteria. The sulfur-bearing
amino acids which naturally occur in the wood have the capability of producing minor amounts
of sulfonic acid rather than sulfuric acid. However, all of these reactions hinge on the presence
of oxygen.
This suggests that our sample was in contact with some small amount of dissolved
oxygen while submerged. This oxygen reacted with the H2S H2S produced by bacteria to form
sulfuric acid, which in turn reacted with additional H2S to form elemental sulfur. Iron sulfides
formed by the reaction between H2S and free iron ions would oxidize immediately and entirely
upon contact with the oxygen available in the water. Any acidification that could be attributed to
iron sulfides occurred before the sample was excavated. This would explain why our XAS
spectra contained no iron sulfides at all, but instead contained large amounts of sulfates and iron
oxides Figure X.
The presence of a large proportion of bound sulfates indicates the originally free sulfates
came into concact with free cations. There is a slim possibility that these free cations were in the
container holding the sample after excavation, or perhaps transferred during handling. The more
likely explanation is that the sulfates came into contact with free cations while still submerged.
Seawater is known to contain many dissolved salts, and the most probable cations are sodium,
magnesium, calcium, potassium, and strontium. Further XAS testing would be reqired to
identify the cation of the ordered sulfate. Once identified, we could determine if the sample is
reacting primarily under water, or if it is being contaminated during excavation and handling.
Our sample had contact with some amount of oxygen while submerged. Other parts of
the ship have undoubtably received different exposure to oxygen due to their position, and
therefore contain different amounts of compounds with the potential to become acidic.
6.0 ACKNOWLEDGEMENTS
I would like to thank Dr. Apurva Mehta for mentoring me on this project, and for
dispensing exceptional advice and encouragement. I would also like to thank Dr. Ritimutka
Sarangi for her generous support and guidance, and for contributing the additional S K-edge
standards used. Thank you to Eleanor Schoefield, for the providing the sample and sharing beam
time. Additionally, thanks to Dr. Steven Rock for his direction of the SULI program at SLAC.
This research was funded by the U.S. Department of Energy, and the Stanford Synchrotron
Radiation Lightsource.
REFERENCES
[1] J. Gillon, " Marine archaeology: Acid attack," Nature, vol. 415, no. 21 Feb, pp. 847, 2002.
[2] D.L. Hamilton, Methods of Conserving Archaeological Material from Underwater Sites, 1
ed. , College Station, TX: Texas A&M University, 1999.
[3] R. Sarangi, J.T. York, M.E. Helton, K. Fujisawa, K.D. Karlin, W.B. Tolman, K.O. Hodgson,
B. Hedman, E.I. Solomon, " X-Ray absorption spectroscopic and theoretical studies on
(L)2[Cu2(S2)n]2+ complexes: Disulfide versus disulfide(1-) bonding," Journal of the
American Chemical Society, vol. 130, pp. 676-686, 2008.
[4] M Sandström, F. Jalilehvand, I. Persson, U. Gelius, P. Frank, I. Hall-Roth, " Deterioration of
the seventeenth-century warship Vasa by internal formation of sulphuric acid," Nature, vol.
415, no. 21 Feb, pp. 893-897, 2002.
[5] S. Kopriva, T. Hartmann, G. Massaro, P. Hönicke, H. Rennenberg, " Regulation of sulfate
assimilation by nitrogen and sulfur nutrition in poplar trees ," Trees - Structure and Function,
vol. 18, no. 3, May, pp. 320-326, 2004.
[6] K.M. Wetherall, R.M. Moss, A.M. Jones, A.D. Smith, T. Skinner, K.M. Pickup, S.W.
Goatham, A.V. Chadwick, R.J. Newport, " Sulfur and iron speciation in recently recovered
timbers of the Mary Rose revealed via X-ray absorption spectroscopy," Journal of
Archaeological Science, vol. 35, pp. 1317-1328, 2008.
[7] B.B. Jørgensen, D.C. Nelson, " Sulfide Oxidation in marine sediments: Geochemistry meets
microbiology," Geological Society of America Special Papers, vol. 379, pp. 63-81, 2004.
[8] M. Sandström, F. Jalilehvand, E. Damian, Y. Fors, U. Gelius, M. Jones, M. Salome, " Sulfur
accumulation in the timbers of King Henry VIII's warship Mary Rose: A pathway in the
sulfur cycle of conservation concern," PNAS, Proceedings of the National Academy of
Sciences, vol. 102, no. 40, pp. 14165-14170, 2005.
FIGURES
Figure 1 Fifteen spectra overlay. (absorption vs energy) This data has not been normalized.
The features at energies greater than the peak at 2482 eV indicate the presence of an ordered
sulfate. This is data (lots of points). Spectra are different samples A,B…
Sample
Edge
Jump
Methionine
C5H11NO2S
Elemental
Sulfur S8
Sulfoxide
R-S(=O)-R'
Sulfonate
RSO2O−
Sulfate
SO42-
sum
R-value
A
B
C
D
E
F
G
H
I
J
K
L
M
N
O
0.6609
0.6897
0.4834
0.2991
0.2558
0.1915
0.2631
0.3627
0.3891
0.3223
0.2596
0.6645
0.3221
0.5338
0.4782
0.0286
0.0249
0.0620
0.1177
0.1623
0.1836
0.1920
0.0912
0.0594
0.1341
0.1475
0.0277
0.0571
0.0371
0.0385
0.0698
0.0832
0.1580
0.2970
0.2828
0.4201
0.4200
0.2787
0.2826
0.4675
0.4350
0.0927
0.2719
0.1264
0.1753
0.0110
0.0211
0.0359
0.0680
0.0964
0.1139
0.1195
0.0674
0.0771
0.1184
0.1212
0.0254
0.0714
0.0284
0.0373
0.0192
0.0608
0.0000
0.0000
0.1598
0.0387
0.0629
0.0643
0.1166
0.0527
0.0516
0.0799
0.1436
0.0069
0.0630
0.8000
0.7150
0.6390
0.4350
0.2507
0.1847
0.1521
0.3976
0.3545
0.1690
0.1744
0.7267
0.3386
0.7000
0.6148
0.9286
0.9050
0.8949
0.9177
0.9520
0.9410
0.9465
0.8992
0.8902
0.9417
0.9297
0.9524
0.8826
0.8988
0.9289
0.011300
0.007700
0.006700
0.002800
0.001600
0.000829
0.001300
0.003500
0.008700
0.001010
0.001350
0.007700
0.004400
0.006700
0.006000
Table 2 The fit fractions, total sum (to 1.0000) and R value are calculated for each location.
Sum is not 1.
Fit
Sulfur
Data
Sulfate
Methionine
Sulfoxide
Sulfonate
Figure 3 Curve fit of the location with the least total sulfur Label axis. Bigger scale
Data is actually lots of points
Sample
Methionine
C5H11NO2S
Elemental
Sulfur S8
Sulfoxide
R-S(=O)-R'
Sulfonate
RSO2O−
Sulfate
SO42-
A
0.0189
0.0461
0.0073
0.0127
0.5287
B
0.0172
0.0574
0.0146
0.0419
0.4931
C
0.0300
0.0764
0.0174
0.0000
0.3089
D
0.0352
0.0888
0.0203
0.0000
0.1301
E
0.0415
0.0723
0.0247
0.0409
0.0641
F
0.0352
0.0804
0.0218
0.0074
0.0354
G
0.0505
0.1105
0.0314
0.0165
0.0400
H
0.0331
0.1011
0.0244
0.0233
0.1442
I
0.0231
0.1100
0.0300
0.0454
0.1379
J
0.0432
0.1507
0.0382
0.0170
0.0545
K
0.0383
0.1129
0.0315
0.0134
0.0453
L
0.0184
0.0616
0.0169
0.0531
0.4829
M
0.0184
0.0876
0.0230
0.0463
0.1091
N
0.0198
0.0675
0.0152
0.0037
0.3737
O
0.0184
0.0838
0.0178
0.0301
0.2940
Table 4 Normalized proportions of sulfur species at each location What does normalized mean?
At what “Total Sulfur” since the fractions depend on that.
What are the units of “Total Sulfur”
Label both axis as “Total”. Give units if possible e.g. gms/cm3
Points come from samples A, B, etc.
Figure 5
FIGURE X