Archaeological Chemistry - American Chemical Society

---

Chapter 9

Downloaded by UNIV OF CALIFORNIA SAN DIEGO on December 17, 2016 | http://pubs.acs.org Publication Date: August 15, 2002 | doi: 10.1021/bk-2002-0831.ch009

Morphology and Microstructure of Marine Silk Fibers 1

2,

Rekha Srinivasan and KathrynA.Jakes * 1

19 Cenotaph Street, Alwarpet, Chennai, India Department of Consumer and Textile Sciences, Ohio State University, 1787 Neil Avenue, Columbus, OH 43210-1295

2

Reports of textiles recovered from marine sites are uncommon and research is needed regarding the physico-chemical structure of water-degraded archaeological textiles and how the marine environment has altered them (1,2). Recovery of trunks full of clothing from the ocean floor site of the shipwrecked S.S. Central America (3) provided a unique opportunity for the study of materials exposed to a marine environment. Recognizing that these textiles could be used to understand both their cultural use in the Gold Rush era and the processes of textile degradation in the deep ocean, research has been conducted on the functions and the structures of the textiles and clothing (4,5), as well on the physical and chemical characteristics of the fibers from which the textiles were made (6-12). Silk textiles appeared to be in fairly good condition apart from some staining and surface deposits. Morphological features revealed through optical and scanning electron microscopy were compared to those of modern silk and silk obtained from historic textiles of comparable age. Elemental composition of the fibers and surface deposits was determined using energy dispersive spectroscopy. Chemical composition and alterations in secondary structure were studied through infrared microspectroscopy. The implications of the stability of silk in this marine site are discussed.

128

© 2002 American Chemical Society

Jakes; Archaeological Chemistry ACS Symposium Series; American Chemical Society: Washington, DC, 2002.

129

Downloaded by UNIV OF CALIFORNIA SAN DIEGO on December 17, 2016 | http://pubs.acs.org Publication Date: August 15, 2002 | doi: 10.1021/bk-2002-0831.ch009

Marine Textile Research Reports of textiles recovered from marine sites are uncommon. Wool and flax fibers from the Mary Rose and the Wasa have been identified employing optical and scanning electron microscopy (13). The fiber content of the sails of the Wasa were identified as "vegetable" but the method for fiber identification was not described (14). Morris and Seifert (75) describe treatment of textiles from the Defence with oxalic acid but do not indicate how they identified the linen, hemp, and silk fibers that they report are present. Conservation treatment of a silk cocade from the Michault was described (16). Peacock (17,18) found that aerated soil at p H 6.5 caused more degradation of silk than unaerated soil at a lower p H . She found surface pitting and fibrillation of the fibers as evidence of degradation and suggested that the greater availability of bacteria and fungi in aerated soil results in enzymatic hydrolysis of the silk fibroin. Water-degraded archaeological textiles showed fibrillation, erosion, perpendicular splitting, and pitting. Fibrillation in silk fibers has been attributed to the beginning of very fine axial splitting called lousiness attributed to prolonged wetness, acid hydrolysis, or boiling (17,18). In 1987 Florian (19) stated that there was no published data on the analysis of silk from a marine site; die only analysis on silk artifacts recovered from wet or frozen sites reported since then has been that by Peacock in 1996 (17,18). N o study of the alterations in morphology and chemistry of silk due to exposure to the deep ocean environment has been reported in the literature. Recovery of many silk textiles from the site of the shipwreck of the S.S. Central America, therefore, provided an opportunity for examination of the chemical and physical characteristics of silk that had been exposed to a marine environment for an extensive period of time. The steamship sank in 1857, and the trunks were recovered in 1990 and 1991 (3,6). Conditions at the site are: depth, 2200 m; solar radiation, nil; pressure, 220 atm; current, 10 cm/s (mean); and visibility, 15 to 20 m (3). Additional values anticipated for other parameters were reported as: temperature, 3 °C; dissolved oxygen, 5.8-6.4 m L / L ; nitrates, 17-23.9 microgram-atoms/L; silicates, 12-25.9 microgram-atoms/L (3). Fibers from the textile artifacts contained within the trunks have been studied using multiple analytical techniques thereby documenting morphological characteristics and elemental composition of deposits (7,10). Study of flax fibers from the trunks in comparison to modern flax revealed lower overall crystallinity of the marine material, and increased crystallite size in comparison to modern flax fibers. Unit cell dimensions were reported to be unchanged (9).

Jakes; Archaeological Chemistry ACS Symposium Series; American Chemical Society: Washington, DC, 2002.

Downloaded by UNIV OF CALIFORNIA SAN DIEGO on December 17, 2016 | http://pubs.acs.org Publication Date: August 15, 2002 | doi: 10.1021/bk-2002-0831.ch009

130 In an attempt to document the effect of the deep ocean environment on textile materials, modern materials were immersed at the site on the ocean floor for subsequent retrieval. While one set of samples was recovered after a three month period, additional samples have remained on the ocean floor since 1991 and are yet to be retrieved. The morphology of cellulosic fibers immersed for a three-month period has been investigated (8). The physical and chemical structure of dyed and undyed cotton fibers from the site compared with those of modern cotton and of cotton immersed on the ocean floor for three months were reported (70,77). Results of preliminary analyses on silk fibers using differential scanning calorimetry, energy dispersive x-ray spectoscopy, and scanning electron microscopy, have been reported (72).

Silk Fiber Structure The filaments of fibroin produced by the larvae of the Bombyx mori (20) provide the silk used commercially for textiles. While the specific amino acid composition of fibroin depends on its source, it contains a large percentage of the amino acids glycine, alanine, serine, and tyrosine. The predominance of these simple amino acid residues, which can pack into a compact structure, is one feature of the chemical and physical structure of fibroin that influences performance properties of the filaments (21). Small amounts of other amino acids including valine, aspartic acid, glutamic acid, threonine, arginine, leucine, isoleucine, histidine, proline, tryptophan, and cystine are present (27). The number of acidic groups on the side chains of this protein is about 2-3 times more than the number of basic groups, affecting such properties as dyeability. The regularly arranged polypeptide backbone is stabilized by hydrogen bonding between the N - H and the C=0 of the amide linkage (22). The commonly encountered conformations of the polypeptide backbone include the a-helix, parallel and anti-parallel P-pleated sheets, and random coil conformations (23). While the a and P forms are found in crystalline regions of proteins and are characterized by regular arrangement of the polypeptide chain, the random coil conformation corresponds to the amorphous regions and is not characterized by the regular packing of the main chain. A transition phase with laterally ordered regions between the crystalline and amorphous phases has been proposed (24,25).

Jakes; Archaeological Chemistry ACS Symposium Series; American Chemical Society: Washington, DC, 2002.

131

Experimental Methods

Downloaded by UNIV OF CALIFORNIA SAN DIEGO on December 17, 2016 | http://pubs.acs.org Publication Date: August 15, 2002 | doi: 10.1021/bk-2002-0831.ch009

Samples To avoid the confounding effect of dye, three textiles (identification numbers 29049, 29054, and 33707) were selected from those recovered marine silks that appeared to be undyed. For comparative purposes, three undyed historic silk textiles (identification numbers 88b, 145b, and 145c) from the period 1860-1880 were obtained from the Historic Costume and Textiles Collection at The Ohio State University. Though the production and usage histories of the marine and historic silks are different, it was assumed that one principal difference between the two sets of silk fibers lies in the fact that the marine silk had been exposed to long term storage in a deep ocean environment, while the historic silks had not. The cumulative effects of 130 years in the two very different environments should be an overriding cause of the most significant differences between the marine and historic silks. Small samples of fabric, ranging in size between 70 and 100 m m , were taken from inconspicuous places in the artifacts (e.g., unfinished seam allowances), so as not to affect the appearance of the garments in future display. Modern undyed and unfinished silk (Style #601) obtained from Testfabrics Inc. (Middlesex, NJ), was used in this work as a reference material and is referred to in this document as Reference Silk. 2

Optical Microscopy A Zeiss Axioplan Research Microscope was used to examine the fibers, employing nominal magnifications of 400 X and 1000 X . Between 10 and 15 fibers of each of the samples were mounted onto glass slides using Permount® (Fisher Scientific), a histological mounting medium with a refractive index of 1.55, and allowed to dry for 24 hours before viewing under a microscope. Images were captured using an Optimas 5.2 image analysis system. Diameter measurements were done on 10 fibers per sample at 5 points along each fiber length. After confirming data normality for each sample, one way analysis of variance ( A N O V A ) was used to test the null hypothesis of no difference between the different sample means. Additionally, Tukey's pairwise comparison procedure was performed to determine the differences between different pairs of means. A l l statistical analyses were performed using Minitab Statistical software, version 11.0.

Jakes; Archaeological Chemistry ACS Symposium Series; American Chemical Society: Washington, DC, 2002.

132

Downloaded by UNIV OF CALIFORNIA SAN DIEGO on December 17, 2016 | http://pubs.acs.org Publication Date: August 15, 2002 | doi: 10.1021/bk-2002-0831.ch009

Scanning Electron Microscopy (SEM) and Energy Dispersive Analysis of Xrays (EDS) In order to investigate fiber morphology, a J E O L J S M 820 scanning electron microscope was used with an accelerating voltage of 20 keV. Analysis of the elemental composition of the fibers and surface contaminants was performed using an Oxford Instruments Group Link Analytical e X L energy dispersive spectrometer with a high resolution Pentafet detector and windowless capability. Fiber samples for both S E M and E D S were mounted on carbon planchettes with carbon tape and sputter-coated with carbon using a Denton Vacuum Desk II coater. Between 5 and 10 individual fibers of each sample were mounted for examination. S E M photographs were taken under a variety of magnifications in order to elucidate various morphological characteristics. When the fibers themselves were being analyzed using E D S , spectra were collected at a magnification of 10 weight %), minor (1-10 weight %), and trace (