Historic Textile and Paper Materials II - ACS Publications - American

Chapter 9

Historic Silk Flags from Harrisburg 1

1

2

M. Ballard , R. J. Koestler, C. Blair, C. Santamaria , and N. Indictor

Downloaded by TUFTS UNIV on November 29, 2016 | http://pubs.acs.org Publication Date: September 28, 1989 | doi: 10.1021/bk-1989-0410.ch009

Department of Objects Conservation, Metropolitan Museum of Art, New York, NY 10028 A group of thirty-four silk samples taken from fourteen brittle Civil War flags was provided by the Capitol Preservation Committee, Harrisburg, Pennsylvania. Elemental analyses of these samples were obtained by energy dispersive x-ray spectrometry (EDS). Total sulfur, ash, and pH of the samples is also reported, and colorants were determined. The presence of mordants, weighting materials, and colorants is discussed with reference to the embrittlement of the silk. The connection between fiber deterioration and color is discussed as well as the effects of contemporary manufacturing treatments on the present fiber condition. None of the samples examined was weighted with inorganic salts. The phenomenon of s i l k fabric degradation i n dark storage from the action of mineral s a l t s used i n the processing of the s i l k , especially the 19th and 20th century use of t i n s a l t s as weighting agents, has been a concern for t e x t i l e curators, conservators and historians for many years(1-4). Standard test methods used for the assay of weighting materials generally requires larger samples than can be s a c r i f i c e d from h i s t o r i c t e x t i l e s . Ordinary scanning electron microscopy (SEM) seems to f a i l as a method for distinguishing weighted from unweighted samples (5), but recently the use of transmission electron microscopy (TEM) on uncoated single fibers has shown what appears to be clear differences (6). As shown e a r l i e r , satisfactory results may be obtained by energy dispersive x-ray spectrometry (EDS) supplemented by ash content determinations. An i n i t i a l study on controlled modern s i l k samples involved the detection of t i n and iron weighting in various recipes chosen form the l i t e r a t u r e (5). With this experience, small samples taken from flags i n the National Museum of American History's D i v i s i o n of Armed Forces History were analyzed (1). The flags dated from the C i v i l war through World War I . The results of analysis showed tin-weighting i n only one of the degraded s i l k flag samples - a French g i f t to General Pershing Current address: Conservation Analytical Lab, Smithsonian Institution, Washington, DC 20560 Current address: Chemistry Department, Brooklyn College, and the Graduate Center, City University of New York, Brooklyn, NY 11210

2

0097-6156/89/0410-0134$06.00/0 ο 1989 American Chemical Society

Zeronian and Needles; Historic Textile and Paper Materials II ACS Symposium Series; American Chemical Society: Washington, DC, 1989.


9. BALLARD ETAL

Historic Silk Flags

135

during World War I . A consistent feature of that sample set vas a sulfur content higher than could be accounted for from the sulfur containing amino acids i n s i l k , and probably higher than any sulfur that may have been part of the colorants. Only cysteine and methionine i n Bombyx mori contain sulfur ( t o t a l S- ca 0.05-0.07%) (7,8). Early synthetic dyes with sulfonic acid groups were indicated by spot tests on some of the colorants; but sulfur was prominent in dyed and undyed samples a l i k e . English, French, and German 19th century dye texts indicate that many s i l k s were not so much t i n weighted as they were 'brightened' by a f i n a l 'sour' or rinse i n s u l f u r i c a c i d . White s i l k tended to be bleached of any o r i g i n a l yellowness with 50^(9-13). Accretions of d i r t , s o i l , or atmospheric pollutants on the surface may also account for some of the s u l f u r . The War Department donated these flags to the Smithsonian I n s t i t u t i o n i n 1919. It i s known that some of the flags held by the War Department were i n poor condition as early as the 1880's; the entire c o l l e c t i o n was shipped to the Smithsonian I n s t i t u t i o n by boxcar (14). In the present study, a well documented c o l l e c t i o n of samples was examined (Table I , Figures 1 and 2). The Capitol Preservation Committee, Harrisburg, Pennsylvania provided samples of some C i v i l War flags, manufactured mostly by Horstmann Brothers & Company or Evans & H a s s a l l . These flags had been sealed i n glass cases beneath i n d i v i d u a l s i l k chiffon sleeves on their hoists form 1901 u n t i l a conservation program was organized i n the early 1980's. In addition to conservation documentation, a f u l l scale h i s t o r i c a l review of the flags has been carried out and published (14). Experimental. EDS Analyses. Sample preparation and data treatment has been described (1,5). Photomicrographs and printouts of EDS scans were retained for f i l e s . Ash and Total Sulfur Analyses. Samples were removed from the specimens and submitted for ash and t o t a l sulfur analysis to Schwartzkopf Microanalytical Laboratory, Woodside, N.Y. 11377. For ashing, samples were burned under oxygen at 900-1000°C for ca one half hour. For s u l f u r , samples were treated with potassium metal to convert sulfur to s u l f i d e s , followed by conversion to methylene blue for spectrophotometric analysis (at 670 nm). Except for very small samples (less than 0.2mg), values for ash and sulfur have an uncertainty of ca ±10% of the reported values. 2 pH and Color Analyses. Samples 1 χ 1 cm were placed i n test tubes with 3-4 ml deionized water. pH of the water was measured after five minutes at ambient temperature using a Corning Model #12 pH meter and a Markson combination electrode. (The pH of modern s i l k samples measured i n the same way was ca 5 . 9 ± 0 . 1 ) . The same samples used for pH measurements were analyzed for colorants according to procedures of H. Schweppe (15,16)



136

HISTORIC TEXTILE AND PAPER MATERIALS Π

Table I . Description of Flags Sampled

Ha.

Raajstratiaa He.

ItefMMNlt

012 019 022 026 096 137 175 066

1985.012 1985.019 1985.022 1985.026 1985.096 1985.137 1985.175 1985.066

Stata Calar 23rtf PA Stata Calar 28th PA 29th PA Stat* Calar 31st PA ( 2 * ras) SiaUCalar 63r4 PA SlaU Calar? 82*4 PA Stata Calar Stata Calar Uakaawa 51st PA Stata Calar

Evaus I Hassall Evaas & Hassall? HarsUaaaa Bras. & Ca. ataMaa Brai. S£a. Evaas Si Hassall? Uakaawa Evaas Si Hassall? ataMaa Bras. * Ca.

167 1985.167 168 1985.168

51at PA 51st PA

Uakaawa Uakaawa

005 1985.005 050 1985.050

59th PA (2a* Cav) SUU Staaëard 44th PA (1st Cav) StaU Staaëarë

Har staiaaa Bras. & Ca. Uakaawa

002 1985.002 103 1985.103

78th PA 67th PA

Uakaawa Uakaawa

Histarical

Gaiaaa, Natlaaal Gaiaaa. Natiaaal

Hatiaaal Raahaiatal Natiaaal RaaJaMatal

Maaafactarar



BALLARD ET AL.

Historic Silk Flags

Figure 1. 1985.002; 78th PA National Regimental Flag.

Figure 2. 1985.019; 28th PA Regiment State Color.


138


Results and Discussion


Table II l i s t s the EDS results of the t h i r t y - f o u r specimens q u a l i t a t i v e l y . Only elements of atomic number greater than 11 are sensed i n the EDS analyses. The results were obtained an normalized percentages using a software program (ASAP, a Kevex proprietary procedure) usually employed i n analyses from t h i s laboratory. The signs used i n Table II should be interpreted as follows: ++ - The element i s present i n s i g n i f i c a n t amounts (more than 30%, of elements of atomic number greater than 11). The element i s found i n a l l scans. + - The element i s present i n a l l scans (more than ca 5%) of elements of atomic number 11). (+) - The element i s present i n most, but not necessarily a l l scans (ca 5% of elements of atomic number greater than 11). - - The element i s not detected. The wide v a r i e t y and v a r i a b i l i t y of elements detected for i d e n t i c a l l y colored s i l k s suggests, as previously observed, that the methods of coloring these h i s t o r i c a l flags was not standardized. Every sample showed the presence of i r o n . Of the thirteen blue specimens eleven contained t i n ; two did not. Of the ten red samples, eight contained t i n ; two did not. Only two of the ten white samples contained t i n . Sulfur, calcium, s i l i c o n , aluminum were found i n a l l samples; chlorine, potassium, and sodium were frequently found; magnesium was detected occasionally i n trace quantities. Table III l i s t s the results of pH measurements, ash, and t o t a l sulfur analyses grouped according to sample color. A l l pH measurements showed lower values, (usually by more than 2.0 pH u n i t s ) , than were found for modern s i l k samples suggesting rather c l e a r l y the occurrence of chemical changes i n the h i s t o r i c s i l k s . Blue samples had the highest pH, average ca 4 . 8 ± 0 . 5 ; red average ca 4.2 ± 0 . 3 ; white (three) ca 4 . 1 ± 0 . 3 . Except for a single example (012) the sulfur analyses were substantially higher than values reported for untreated s i l k , i . e the sulfur accountable to the presence of cysteine or methionine i n the f i b r o i n chain of s i l k . Values obtained for t o t a l sulfur ranged from ca 2 - 9 times the value expected from untreated s i l k . The single low value found for a blue sample was very close to that expected for untreated s i l k ; the highest value was for a white sample. Most of the samples (twenty-five of thirty-four)had a sulfur % of 0 . 3 5 ± 0 . 1 2 , ca 5 times the amount present i n untreated s i l k . The ash content of the samples indicated c l e a r l y that none of the samples was weighted with inorganic substance. The clearest property d i s t i n c t i o n among the colors was observed i n the ash % r e s u l t s : The blue samples gave the highest values, average ca 4.9±0.4%; red, average ca 3.2±1.0%; white, average ca 1.6±0.3%. These average values were calculated neglecting data from small samples (Table I I I , date with * ) . Although there were a few overlapping r e s u l t s , for the most part there seems to be a high c o r r e l a t i o n between color and ash content. The colorants for a l l the flags was found to be consistently cochineal for the reds and Prussian Blue for the blues. Although the


9.

139

Historic Silk Flags

BALLARD ET AL.

Table I I . Qualitative EDS Results «a.»

Hi

012B V Β

•


019S V Β

(•)

022R (•) V (•) Β (•) Q26R (•) V Β (•>

-

096R (•) V (•) Β (•) 137Β V Β

S

Cl

Κ

Ca

Fa

(•) (•) (•) (•)

4

4 • •

(•) (•) (•)

• 4-

4 • 4

4 (•) 4-fr

(•> (•) (•>

4 • •

• •t-t4

-

(•)

4

-

-

4• 4-

4-fr

(•) (•) (•)

• •

• • •

_

(•)

-

4· 4· 4-

4» 4· 4-fr

•

+

•

-

4·

4-

4-

•

•

-

4·

4·

4-fr

+

+

4-

•

•

4-

•

4-

(•)

4-

4·

4~fr

4-fr

• • •

4• 4-

(•)

• 4-

4 44

4· 44-fr

4--fr

4-

•

• •

• 4»

4· 4

•

(•)

-

M

4• 4-fr

4-fr

•

•

(•)

_

*

4-

4-fr

(•)

4

(•)

4-fr

4-fr

(•) • (•) • (•)

4· 4-fr

-

• 44·

(•) 44-

4-fr •

(•) (•) (•) (•)

• • •

+ 4-fr 4-fr

• 4· (•)

• 4 4· • 4-fr (•)

4

(•> (•) •

• • • •

• • + •

-

(•) (•)

(•)

-

17SB + V Β (•>

-

_

_

-

OMR (•) ¥ (•) Β (•)

_

167R (•) V Β

_

-

168Β V Β QQ5B 090Β 0Ο2Β 103Β *RVB 4-fr * « (4) •

-

•

+

• • •

_

-• •

Sa

Si

ΑΙ

-

•

•

-

4-

-

_

(•) W (•)

_

-

4(•)

(•)

(•)

(•) 4-

*

4-

4-fr 444-

_

-

4-fr 444-fr •

-

4-fr

-4 -

-

-

4· 44 4-

rwfwrs ta aatar af laapaj r

Fraaaat ia atjatftaaat aaiiaatf ( > 3 0 « af akawats af ataaria aa. >11). Praaaat. (aj.ahSOfi af akawats af ataaria aa. >1t) Aaaaat (aat aVtaataaO. Pasrtaaj arasaat («J. 9S ar toss af alaawats af ataaria aa. >11).



Table I I I . pH, S u l f u r , Ash of H i s t o r i c Flags tapi»

pH»

Mfw/* « f

c

A*fc i


BUKMrtPta 012 019 022 02ft 096 137 173 066 167 168 089 098 002 108

4J61 4.97 9.10 430 4J06 4J64 4.17 330 4.19 4J00 4JB1 930 4JB9 939

0371 0.27 0.19 036 030 OJB 032 0.19 030 034 030 0.10 039 030

9.99 6.49 732 4.75 3J62

Historic Textile and Paper Materials II - ACS Publications - American

Recommend Documents