Deciphering Ageing Effects in Green-Dyed English Wool Carpet Yarns from the 1840s

Abstract

In 1842, carpet manufacturer W.H. Worth of Kidderminster, England, began assembling a sample book of wool yarns dyed with natural dyestuffs. This paper reports on a study of the “Greens” section, which contains sixteen yarn samples—six still green and ten now ranging from tan to dark brown. The accompanying recipes list similar ingredients: old fustic and either “mixture” or extracet of indigo. To verify whether Worth’s recipes were followed, the yarns were analyzed using HPLC-DAD and FORS. Additionally, mock-ups were prepared according to Worth’s green dye recipes and subjected to thermal ageing to explore potential causes of discoloration. Preliminary analysis of the historic samples revealed that the discoloured yarns contain both indigo and indigo carmine, while the still-green samples contain only indigo carmine. This suggests that one or more components of the indigo vat may have contributed to discoloration. To test this hypothesis, contemporary wool yarns were dyed using a Worth green recipe, with and without indigo, at varying pH levels. These were thermally aged, and their colour changes monitored. HPLC-DAD and FORS analyses of the mock-ups were compared to the historic samples to identify dyeing conditions that may have led to the observed browning.

1. Introduction

Carpet manufacturer William Henry Worth of Kidderminster, England, began in 1842 to assemble a sample book of wool yarns arranged according to colour (Figure 1). Recipes accompany more than two-thirds of the 183 samples in his book [1]. Consistent with the date of the sample book, the recipes call for the use of natural dyestuffs. Most of the yarn samples still display the probable original colour. However, the yarns in the Greens section of the book are a very noticeable exception to this colour stability; only six of the sixteen “Green” samples are still green (Figure 2a). The others range in appearance from tan to dark brown (Figure 2b).

All the recipes in the Greens section list both the natural yellow dye fustic and a blue dyestuff, in addition to alum. Those recipes accompanying samples that are still green call for “Extract of Ind.” (=extract of indigo) which is a recognized alternative name for the blue dye indigo carmine (Figure 3a) [2]. Most of the recipes accompanying the discoloured yarns do not mention a particular blue dyestuff. Instead, they call for “Mixture” (Figure 3b). Only two of the discoloured yarns list “Extract of Ind.”. Nowhere in the book does Worth list the composition of “Mixture” or describe how it was prepared. There is essentially no correlation between the ratio of blue dye to fustic in the recipes and the extent of browning. Samples with nearly identical recipes display completely different hues; for example, one appears silvery white and another on the same page with similar ingredient ratios is dark brown. Also, of the two samples with recipes calling for vitriol, one remains a brilliant green, whereas the other has degraded to brown.

Old fustic contains three major dye compounds: morin, maclurin, and kaempferol. Both morin and maclurin are soluble in a basic aqueous solution, which will turn brown when exposed to air [3]. It is likely that the browning occurs in the old fustic components but is driven by environmental factors. The primary difference between the recipes is the identity and proportion of the blue dye, suggesting that an interaction between component(s) in the blue dyestuffs and in the fustic may have contributed to the discoloration in some yarns.

In A Manual of the Art of Dyeing (1853 3rd edition), James Napier advises that “when the cloth is dried from the sulphate of indigo solution, the acid of the chemic must be neutralized… were the acid property to prevail in the least, it would destroy the yellow upon the cloth to be dyed green; and were the alkaline matter predominant, it would brown the yellow, and the green would assume a blackish olive shade” [3]. This comment suggests that poor pH adjustment of the dyebaths could potentially explain the browning observed in Worth’s greens.

A previous study [4] on carpet samples produced by the same manufacturer provided additional insight into the colour stability of Worth’s dyestuff. These samples, which are part of the V&A collection today, and which were made in the mid-19th century, possibly intended for the Great Exhibition, still retain their original green and yellow hues, in contrast to the discoloration observed in the yarns from Worth’s sample book.

To elucidate the disparities in the current appearances of the historic yarns, small samples of all Worth’s “green” yarns were analyzed by High Performance Liquid Chromatography with a Diode Array Detector (HPLC-DAD), UV/Vis reflectance spectroscopy and fluorescence spectroscopy. Wool yarn mock-ups were prepared, following each of Worth’s recipes (

Table 1. Matrix of the Worth book receipts.

Green Yarn Sample		Dyes Component			Mordant
		Old Fustic	Mixture	Extract of Indigo	Alum	Vitriol
WHW 39T 1		X	X		X
WHW 39B		X	X		X
WHW 40T		X	X		X
WHW 40B		X	X		X
WHW 41T		X	X		X
WHW 41B		X	X		X
WHW 42T		No Text	No Text	No Text	No Text	No Text
WHW 42B		No Text	No Text	No Text	No Text	No Text
WHW 43T		X		X	X
WHW 43M		X		X	X
WHW 43B		X		X	X	X
WHW 44T		X		X	X
WHW 44B		X		X	X	X
WHW 45		No Text	No Text	No Text	No Text	No Text
WHW 88		Number 16 Green 120 lbs of Worsted

¹ WHW = William Henry Worth; the number indicates the sample book page, T = top sample, M = middle sample, and B = bottom sample.

). The effects of varying the pH of the dyebath were observed, starting with a recipe that accompanies a Worth sample that is still green. After the dye components of Worth’s “Mixture” were determined to be indigo carmine and indigo, yarns were also dyed using this mixture.

Portions of all the contemporary dyed samples were thermally aged, and colour changes were noted. None of the treatments applied to the contemporary yarns sufficiently reproduced the discolouration noted in Worth’s samples. Possible explanations for the lack of significant colour changes in the aged mock-up samples, as compared to the Worth yarns, are suggested.

2. Materials and Methods

2.1. Materials

Undyed wool yarn was purchased from Brush Creek Woolworks, Colorado, USA. Indigo carmine, morin and potassium hydrogen tartrate were purchased from Acros Organics, Morris Plains, NJ, USA and indigo from Sigma-Aldrich, St. Louis, MO, USA. Fustic was purchased from Kremer Pigments, New York, NY, USA. Maclurin was purchased from Fisher Scientific, Fair Lawn, NJ, USA and kaempferol from Alfa Aesar, Ward Hill, MA, USA. Aluminum potassium sulfate was recrystallized from material originally purchased from J. T. Baker, Phillipsburg, NJ, USA. Calcium hydroxide was obtained from Alfa Aesar, Ward Hill, MA, USA and sulfuric acid, Optima grade, from Fisher Scientific, Fair Lawn, NJ, USA. Water was purified to a resistivity of 18 MOhms.

Samples less than 0.5 cm in length were removed from each of the yarns in the Greens section of Worth’s book for spectral and chromatographic analyses.

2.2. Methods

A series of mock-ups was prepared to replicate the processes and reagents Worth may have used, and the resulting yarn samples were aged in a thermal ageing oven to simulate degradation over time. Three sets of mock-ups were prepared.

To prepare contemporary, undyed wool yarn for the dyeing of mock-up samples, yarn was first pre-mordanted in an aqueous solution containing 1.5 g of aluminum potassium sulfate and 0.286 g of potassium hydrogen tartrate in 750 mL deionized water. The solution was heated and stirred in a Pyrex beaker to dissolve the solutes. Approximately 10 g of untreated yarn was added, the contents brought to a boil, and boiled for 6 h. The mordanted yarn was removed from the beaker and hung in a fume hood overnight to dry.

Thirty-centimetre lengths of the alum-mordanted wool yarn were dyed in Pyrex beakers containing alum (aluminum potassium sulfate) and dyestuff(s) following the proportions in each of the Worth recipes in the Greens section of his book. All the listed ingredients were boiled together for three hours. Indigo carmine was used as the dyestuff for both “extract of indigo” and for “mixture” at this stage of the investigation.

Control yarns were prepared with each of the dyestuffs indigo carmine and old fustic. Yarns were also dyed with the individual components of old fustic—morin, kaempferol, and maclurin—following the same procedures.

The indigo bath was prepared by adding 1 g of indigo powder, 1.75 g of ferric sulfate and 2.375 g of calcium hydroxide to 100 mL of ultrapure water. The mixture was heated with stirring for approximately four hours. The solution was allowed to cool and settle in a capped Pyrex jar for 72 h, during which time the contents formed distinct layers. Aliquots were taken from the yellowish middle layer, between floating solids on top and a greenish precipitate [4].

Wool yarns dyed with fustic were subsequentially dyed using mixtures of a prepared indigo bath and an indigo carmine solution in order to observe any effects that the components of an indigo bath typical of the period—indigo, ferrous sulfate, and calcium hydroxide—might have had on Worth’s dyed yarns [3]. Indigo bath/indigo carmine solution dye ratios of 2:1, 1:1, and 1:2 (vol:vol) were used.

For investigation of the effect of dyebath pH, the Worth recipe accompanying still-green sample 44T was chosen, and indigo carmine was used as the dyestuff. Yarns were dyed in the bath as described above, which had a pH of circa 3.5, and also in baths prepared following the same recipe but with pH adjusted to 5, 7, 9, or 11 using a calcium hydroxide solution and sulfuric acid.

Mock-up samples were thermally aged in a Weiss WKL34 oven for a total of 111 days. Initially the oven was set at 37 °C and 70% relative humidity (RH) for five days. The RH was then lowered to 65% at 37 °C and maintained at this setting for two weeks. Lastly, the oven settings were adjusted to 70 °C and 65% RH and maintained there for 55 days. At this time visible changes in the appearance of the mock-up samples were observed, and cuttings were taken for testing.

All historic and mock-up samples were analyzed using the same procedures. The techniques used were UV/vis reflectance (Agilent Cary 60 spectrometer, Santa Clara, CA, USA) and fluorescence spectroscopies (Agilent Eclipse fluorometer), and HPLC-DAD (Agilent 1100 series) [4] (see Appendix A for experimental details).

3. Results

The contemporary wool yarns dyed according to the Worth recipes displayed a wide range of hues, from bright aqua to golden yellow, and included olive-green yarns (Figure 4). None of these yarns have the brown appearance of the discoloured Worth yarns. Also, it should be noted that none of the mock-ups are a bright green, and none closely resemble the hues of the Worth yarns that remain green.

3.1. Spectroscopy

The few Worth samples that are still green displayed a long wavelength absorbance maximum at approximately 623 nm and a band in the blue with a maximum between 425 and 431 nm. This is exemplified by the FORS data obtained for Worth sample 44T (Figure 5a). These spectral features are similar to those of the indigo carmine and the yellow dyes, respectively. The second derivative spectrum was taken to pinpoint the locations of the absorbance maxima, which appear as sharp minima in this spectrum (Figure 5b).

In contrast, FORS of all Worth yarn samples which had turned brown had a major absorbance band in the red region of the visible spectrum with maxima between 623 and 632 nm (±4 nm) (Figure 5c,d). All the spectra exhibited one or more absorbance bands in the blue, with the longest wavelength band in this region at or above 440 nm.

The mock-up yarn samples dyed following the Worth recipes all displayed a strong fluorescence with characteristics due to components of fustic dyestuff. This is demonstrated for the mock-up sample dyed according to Worth recipe 44T (Figure 6a,b). However, the discoloured Worth samples showed negligible fluorescence. A weak fluorescence was detected from only the Worth samples that are still green. This is illustrated for sample 44T in Figure 6a,c).

3.2. Variations in the Dyebath Recipes

The effect of varying the dyebath pH on mock-up yarns dyed otherwise following Worth recipe 44T can be seen in Figure 7. Increasing the alkalinity of this dyebath led to an increased brownish tint to the dyed yarns. The most basic bath (pH 11) produced an olive-green yarn that somewhat resembles the discoloured Worth samples.

The yarns dyed in a bath with alum, fustic, and mixtures of an indigo carmine solution plus aliquots from an indigo dyebath ranged in appearance from olive-brown to green. The more indigo carmine, the greener the resulting dyed yarn appeared (Figure 8).

3.3. HPLC-DAD

Samples of contemporary indigo carmine and old fustic were analyzed by HPLC to confirm the elution times of the components of these dyestuffs. The results are shown in Figure 9a and Figure 9b, respectively. For indigo carmine (Figure 9a), the largest peaks in the chromatogram that was recorded at 610 nm eluted at about 2.3 min and 12.3 min. These peaks are due to the disulfonated dye; the peak at 13.6 min may be monosulfonated indigo [5]. The broad band eluting at about 7 min has also been observed by other dye researchers but has not been identified [5]. As can be seen in Figure 9b, the prominent peaks corresponding to major components of old fustic all elute between 12 and 17 min. Morin appears at about 13.2 min and kaempferol at 16.8 min.

Figure 10a,b show the HPLC-DAD chromatograms recorded at 610 nm and 360 nm, respectively, for extracts of the Worth yarns 40T (now brown) and 44T (still green). The 610 nm chromatogram for sample 44T clearly resembles that for indigo carmine. In contrast, the major component in the extract of discoloured yarn 40T is a degradation product that elutes at approximately 11.8 min, and the major indigo carmine peak at approximately 2 min is barely detectable. Comparison of the chromatograms recorded at 360 nm for yarns 40T and 44T indicates that the relative amount of morin, a major yellow dye component of old fustic, is markedly reduced in the yarn that has turned brown. The peak for the indigo carmine degradation product is as large as that for the morin peak in this sample. In contrast, the HPL chromatograms for all the mock-up samples displayed peaks for the unaltered dyes.

After more than three months of thermal ageing, the mock-up yarns displayed slight changes in their appearances. The extent of these changes depended on which dyes were present, as seen in Figure 11. The sample of undyed yarn on the left is included for comparison. The eleven Worth mock-up samples acquired tannish or brownish tints to varying extents (e.g., green arrow in Figure 11). The yarns dyed with indigo carmine faded (blue arrow) and yarns dyed with fustic or its components darkened (brown arrow).

Figure 12 illustrates the colour changes that occurred upon thermal ageing of the yarn samples dyed in baths of various pHs. All of these samples became distinctly brown; the depth of the brown shade increased slightly with the acidity of the dyebath (Figure 12). When the yarns dyed with the different ratios of indigo/indigo carmine were thermally aged, they also browned. The extent of browning was somewhat greater, the higher the relative amount of indigo carmine in the dyebaths (Figure 13). The mock-up yarns dyed with any combination of dyestuffs that included both indigo and indigo carmine became more brown when thermally aged, than the yarns dyed with indigo carmine as the only blue dye.

4. Discussion

When Worth’s recipes for dyeing wool yarn green were followed—using indigo carmine as the blue dyestuff component—only a few of the dyed yarns actually looked green, and these yarns were distinctly olive-green. The laboratory-dyed yarn that appeared the greenest was dyed in a bath consisting of a mixture of one part of the appropriate layer from an indigo vat and two parts indigo carmine solution. Worth’s book does not include any recipes for this particular ratio of blue dyestuffs. However, the analysis did show that the “Mixture” he used was a combination of these two dyes in proportions that could not be established. These results suggest that by avoiding an indigo vat in his dyeing processes, Worth and/or his dyer accepted that the green colours he produced for his carpet yarns would not include what might, today, be considered a true green.

The lack of fluorescence in the discoloured Worth yarns implies that the fluorescing components in the original dyestuffs have degraded. These components are very likely to be the flavonols morin and kaempferol, two of the major components of old fustic. These dyes have been observed to brown as they degrade in solutions exposed to air [6,7].

The appearance changes in the thermally aged mock-up yarns indicate that when dyeing wool yarns green in a single dyebath with old fustic as the yellow dyestuff, use of the disulfonated indigo carmine is less likely to result in subsequent browning of the yarn over time than if an indigo vat is used as the source of blue dye. However, longer oven ageing times might have led to more marked colour changes. Also, discolouration of the wool yarn itself may possibly have contributed to the browning observed [8]. The lack of detailed knowledge of the conditions under which the sample book was stored over its almost two-century history has unavoidably prevented a precise re-creation of the ageing conditions for the mock-up samples.

Additional challenges to re-creation of the effects of a centuries-long ageing time on the appearance of the dyed yarns also were confronted. Previous XRF measurements on the Worth carpet yarns and the sample book paper revealed the presence of the heavy metal lead and of arsenic [1]. These results strongly suggested that the book had been treated with inorganic pesticides sometime in the past. In the 1920s pesticides containing these poisons were very popular for protecting wool fibres from pests. Such treatment about a century ago may have contributed significantly to the apparent very good physical condition of the book and yarn samples [1]. A short report describing the effect of pesticides on wool samples dyed with some natural dyes suggests that treatment with these repellants darkened the yarns, including those dyed with indigo or old fustic [9,10].

All the recipes in the Worth book appear to be intended for a one-pot dyeing process. There is no indication that yarns were to be lifted from a dye-pot containing the yellow dyestuff old fustic, dried, and then introduced into another dyebath containing a blue dye to create a green-coloured yarn. Instead, the green recipes listing alum mordant, old fustic, and extract of indigo or “mixture” imply that these ingredients were all put into one dyebath for dyeing the wool yarns green. It is likely that these dyebaths did not contain enough acid to ensure that the resulting dyed yarns would remain acidic over time. As a result, the yellow colours of the fustic dye components have degraded to brown.

It is possible Worth eventually realized this problem when he used old fustic as the yellow dyestuff in his recipes for green wool yarns, and that he needed to ensure the dyed yarns were not alkaline if he wanted his product to remain green. The last recipe in the greens section of his dye book specifically calls for the addition of “1/2 pint of Vitriol” (= sulfuric acid) to the dyebath. The samples accompanying this and the other recipe on this page are the only ones that are still green. In the former case, the presence of the sulfuric acid in the dyebath has ensured that this yarn remained sufficiently acidic; thus the yellow fustic components have not undergone alkaline-induced browning. An explanation for the stability of the other sample on this page has not been confirmed.

Unlike all the other samples in the book, those on one page of the Greens section do not appear to match their recipes at all. Of the three yarns on this page, the top sample is now tan even though the recipe calls for extract of indigo rather than mixture. The middle yarn appears not to have been dyed, and the third yarn has turned a dark brown despite a small amount of “vitriol” being listed in the accompanying recipe. HPLC provided explanations for each of these observations. The top yarn contained only old fustic, no blue dyestuff, and there were no colourants in the middle yarn at all. The dark brown sample at the bottom of the page was found to contain roughly equal amounts of indigo carmine and its degradation products; the presence of these dyes was dwarfed by huge amounts of morin and old fustic degradation products. The small amount of sulfuric acid listed in this recipe was not adequate to prevent browning of the very large amount of fustic. Curiously, there is a notation in the margin of this page, “Three Greens which work in British”. This type of comment has not been added to any other recipe in the book.

The Worth company was still producing green yarns with a mixture of old fustic and indigo carmine for use in carpet manufacture almost two decades after Worth started his sample book [4]. These carpets, now in the collection at the V&A, were likely made in connection with the Great Exhibition. Notably, they remain in very good condition, with their green yarns still vibrant, except for one sample that may have been damaged by light. This suggests that factors beyond the dyestuff composition itself—such as environmental conditions, yarn treatments, or mordanting techniques—may have played a role in the browning of the Worth green samples. The contrast between the well-preserved green hues in the carpets and the degraded greens in the sample book, which was exposed to pesticide treatments and possibly to fluctuating environments, highlights the significant impact of external factors on colour stability.

Analyses of the yarns in the Greens section of Worth’s yarn sample book confirmed that he used the less expensive, less stable yellow and blue dyes listed in the recipes that accompany these samples. He may have chosen these dyes very intentionally, lowering costs and thus keeping down the prices for his products, because he was confident that the carpets manufactured with his colourful yarns would be carefully protected from exposure to daylight and moisture in Victorian homes.