JAIC 1989, Volume 28, Number 2, Article 4 (pp. 97 to 115)
JAIC online
Journal of the American Institute for Conservation
JAIC 1989, Volume 28, Number 2, Article 4 (pp. 97 to 115)


Janet E. Miller, & Barbara M. Reagan


2.1 Fabrics

THE NATURALLY AGED, UNDYED, HISTORIC SILK fabrics used in this study were selected from samples obtained through letters of request sent to major museums, universities, historical societies, and private collectors in the United States. The 54 undyed, historic silk samples evaluated were between 20 and 400 years old and included both unweighted and weighted silk fabrics. Among the historic silk fabrics evaluated, 49 contained cultivated (Bombyx mori) silk, including 21 plain weaves, 10 satins, 12 pattern weaves, 2 crepes, 2 twills, and 2 stockinette knits. The remaining 5 fabrics were plain weave pongees containing wild (Tussah) silk. Both longitudinal and cross sectional slides of silk fibers from each fabric were examined under an optical microscope to identify and classify silk type. Wild silk fibers exhibited lengthwise striations and wedge-shaped cross-sections, compared to the smoother surface and more triangular cross-sections of cultivated silk fibers. The specific histories of the silk samples, except for the approximate age, were largely unknown. However, it was assumed that manufacturing methods, use, care, and storage conditions were typical. The new silk fabric used for comparison was an unweighted, undyed, 160 160, plain weave silk habutai (style #604, Test-fabrics, Inc.) constructed from 30 denier, multifilament yarns. Silk style #604 was degummed after it was woven.

2.2 Characterization of Historic Silk Fabrics

The general condition of the historic silks was classified as good, fair, or poor, based on the number of breaks, snags, and stains and whether it appeared to be strong enough to be displayed without reinforcement. Also determined for each sample were fabric count (yarns per inch) and warp yarn twist and denier according to the test procedures in ASTM D 3775–85, Standard Test Method for Fabric Count of Woven Fabric; ASTM D 1423–82, Standard Test Method for Twist in Yarns by the Direct Counting Method; and ASTM D 1059–83, Standard Test Method for Yarn Number Based on Short-Length Specimens, respectively.1

The historic silk fabrics were subdivided according to age and weighting groups (unweighted and weighted). The following nine age groups were established: 1 (new silk habutai), 2 (1960–75), 3 (1940–59), 4 (1910–39), 5 (1890–1909), 6 (1790–1889), 7 (1700–89), 8 (1600–99), and 9 (1500–99).

The presence and type of weighting agents in the historic silk fabrics were determined by neutron activation analysis (NAA). Specimens (2.5 cm 5.0 cm) were weighed, placed in a Triga Mark II nuclear reactor, then analyzed using a germanium-lithium gamma detector connected to a Canaberra 8180 multichannel analyzer. Weighting standards were prepared for nine water soluble, metallic salts (SnCl2, CuSO4, Al(SO4)3, K2Cr2O7, FeSO4, InCl3, BaN2O6, ZnSO4, and AsCl3), pipetted onto samples of the silk habutai, then irradiated under conditions identical to those used for the historic silk samples. A computer program developed by Higginbothem5 was used to compute the approximate concentration of the metallic elements in the historic silks.

2.3 Methods of Analysis

2.3.1 Yarn Tenacity and Elongation

Strength and elongation tests are used widely for assessing fiber degradation in textile substrates. Thus, yarn-breaking tenacity and elongation tests were performed on the new silk habutai and historic silks to assess the effects of aging on fiber properties. Yarn-breaking load was measured on a Scott Model CRE, following the procedures in ASTM D 2256–80, Standard Test Method for Breaking Load (Strength) and Elongation of Yarn by Single Strand Method,1 except that a multiple-end test method consisting of five warp yarns per test was used. Ten multiple-end tests were performed on each sample. Mean yarn tenacity (based on yarn breaking load and denier) in grams of force per denier and percentage elongation at break were then calculated.

2.3.2 Dilute Solution Viscosity

Viscosity measurements also are used for detecting degradation in textiles. Polymer chain scission and bond breaking at side chains often results in appreciable decreases in viscosity. In this study, the intrinsic viscosities of the new and historic silk fabrics were determined, following the lithium bromide silk viscosity test (SNV 95595-1963) developed by the Swiss Standards Association.14 Specimens weighing 0.100 g each were treated in small test tubes with a saturated lithium bromide solution, then placed in a Precision Scientific laboratory oven for three hours at 60C. After cooling, the silk-lithium bromide solutions were diluted with 5 ml of distilled water, filtered through an ASTM 40–60 fritted glass filter, and then placed in Cannon semimicroviscometers. After temperature equilibrium was reached in a constant temperature bath (20.C), solution flow times were recorded. The average of three effluent times per specimen was used to compute intrinsic viscosity or viscosity number.

2.3.3 Amino Acid Analysis

Seventeen amino acids have been identified in silk fibroin; however, their specific ratios are dependent on the species and strain of silk worm and environmental rearing conditions, processing, use, age, and deterioration. The relative amounts of amino acids present influence the physical properties of the fiber. In this study, the amino acid content of the new and historic silk specimens was evaluated using a Dionex D 300 microbore, single column, amino acid analyzer with a Fisher recorder and a Columbia Scientific integrator. After the silk was hydrolyzed in a 3.0 N p-toluenesulfonic acid bath at 100 C for 30 hours, the amino acids were separated using high-pressure liquid chromatography (HPLC) with a five buffer, stepwise, ion exchange system, and then detected colorimetrically as ninhydrin derivatives.

2.3.4 Optical and Scanning Electron Microscopy

The surface characteristics and fiber fracture patterns in the historic silks were evaluated from longitudinal and cross sectional fiber slides viewed at 430x magnification with an American Optical trinocular light microscope. Photomicrographs were made of selected examples of fiber fracture patterns using an ETC Autoscan scanning electron microscope (SEM). The specimens were prepared by coating the fibers with a thin layer of carbon, followed by a thin layer of gold-palladium.

2.3.5 Photoacoustical Infrared Spectroscopy

An IBM Infrared Spectrophotometer, Model 98, with a photoacoustical accessory was used to record the IR spectra for selected samples of the new and historic silks. In order to eliminate water from the silk specimens, they were stored in a desiccator for one week prior to analysis and then tested in a nitrogen environment. A computer-assisted, subtraction technique was used to detect differences resulting from the experimental treatments. This technique allows for fibers or fabric to be analyzed without grinding the silk into powder to produce powder pellets.16

2.3.6 Statistical Analyses

The silk fabrics were subdivided into nine age classes and two weighting groups (weighted and unweighted). Regression and orthogonal trend analyses were performed on the data for tenacity, elongation, and viscosity using general linear model (GLM) least square tests because of the unequal cell sizes. Sample means were plotted to determine changes over time and with weighting. The analysis of variance test was used to determine significant differences in the amino acid contents among the naturally aged and new silk samples.

Copyright 1989 American Institute for Conservation of Historic and Artistic Works