Home Page Discussion

Salon du Tapis d'Orient

The Salon du Tapis d'Orient is a moderated discussion group in the manner of the 19th century salon devoted to oriental rugs and textiles and all aspects of their appreciation. Please include your full name and e-mail address in your posting.

Mini-Salon 28: Chinese Dyes

by Francesca Fiorentino

There is an interesting discussion on Chinese natural dyes, Logwood dyes in Chinese Rugs, in the Turkotek Archives. I'd like to contribute a report of a recent fascinating exhibition related to it, Intrecci cinesi. Antica Arte Tessile. XV-XIX secolo (October, 12-December 20, 2011) held by Moshe Tabibnia Gallery in Milan, and some related new findings about Chinese traditional dyes.

The Tabibnia Gallery collected and  displayed a number of classical Chinese carpets:
Twenty-six Ningxia, from the 15th  to the beginning of the 19th century;
Five Peking, from the second half of the 16th to the first half of the 17th century;
Some antique Chinese silk textiles;
Seven Gansu rugs, and
Twenty from the Tarim basin.

Details of the technical analyses and comprehensive artistic, historical and scientific articles are included in the exhibition catalog, Intrecci Cinesi. Antica Arte Tessile. XV-XIX secolo (Moshe Tabibnia, Milano 2011). I had the pleasure of hearing directly from Moshe Tabibnia why he decided to investigate the actual appearance of antique Chinese rugs. In 2009 he came by a "dragons" chair cover with a red field, which is unusual for a classical Chinese carpet.

Catalog number. 22: Chair cover with dragons, Ningxia, China, ca. 1700. Wool pile on wool and cotton foundation; 67 x 76 cm (Gallery Moshe Tabibnia, Milan)

Carpets with bright red fields and traditional designs were known only in early Chinese literature and some paintings; there are no extant specimens. Field colors usually range in a wide variety of yellow tones such as pinkish, brownish and reddish.  This chair cover was the first actual specimen with a red field, although they were hypothesized to exist by scholars. This discovery suggested that it would be useful to investigate the dyes that were used by the Chinese.

As much information as possible about the traditional colors and dyes were collected from a selected bibliography (Catalogue page 80). Chinese dyers had remarkable skills that allowed them to obtain refined hues, using dye-baths with different colors to provide the desired tone, after which the yarns were exposed to open air and light to fade and reach the final shades. According to the published information, the preferred tone of red was strawberry, but many reddish tones (pink, peach, apricot, salmon) were obtained from a yellow bath plus a fading red bath. The family of "redwoods" provided this kind of dye: sandalwood (Pterocarpus santalinus), logwood (Hematoxylum campechianum), but most often brazilwood and its most widespread variety in the far east, sappanwood (Caesalpina sappan). Other red shades came from common madder (Rubia cordifolia) pure or mixed with yellow dye, beach mulberry (Morinda citrifolia) and occasionally from organic molecules such as crimson lake (Kermes) and Indian lake (Kerria lacca). Purple came from roots of gromwell (Lithospermum erythryorizon). For yellows, there were vegetal sources, some still unknown: dyer's weed (Reseda luteola), gardenia (Gardenia jasminoide), pagoda tree (Sophora japonica), Amur cork tree (Phellodendrum amurense), some Eurasian smoke trees. Less often, turmeric (Curcuma longa), saffron (Crocus sativus),  tangerine and Chinese rhubarb were used. From safflower (Carthamo tinctorius) came carthamon yellow and red-purple carthamin. Blue was obtained from Chinese indigo (Poligonium tinctorium) and woad (Isatis tinctoria). A darker tone came mixing indigo and brazilwood. Green resulted from indigo plus pagoda tree yellow and by buckthorn (Rhamnus). Black and brown tones were obtained with tannin dyes such as walnut, gall nut plus ferrous sulphate, or logwood.

Moshe Tabibnia's aim of creating a database of natural dyes organized by geographical areas and ages was supported by scholars from the University of Milan and the University of Bergamo (1), whose collaboration provided important scientific contributions to the questions of dye identification, color fading and color changes. The  thorough scientific analyses applied to 14 Ningxia  and Peking specimens focused on field colors including yellow, orange, peach, apricot, salmon, beige, brownish pink and brownish yellow and on some secondary colors including lighter yellow tones, in order to compare the chemical sources. The analyses used three techniques, depending on the yarn color: High performance liquid chromatography (HPLC), surface enhanced Raman spectroscopy (SERS), and visual band reflectance spectrometry (Vis-RS).

HPLC is often used to identify dyes in antique textiles; SERS on silver colloid has been applied to textiles more recently; Vis-RS has rarely been used to study textiles. HPLC and SERS are invasive techniques and require yarn samples (1-1.5 cm. in length), Vis-RS is non-invasive and very fast, although for some dyes the results must be confirmed by other chemical analyses (when the spectrum recorded is not already known or when there is not enough dye to record a spectrum.
Using Vis-RS, the pile yarns can be studied on the back, where colors tipfade the least, and on the front, at the knot basis as well as on the top of the pile. This makes it possible to assess color degradation or fading, and even to record changes in the spectrum curve for each dye analyzed and therefore be able to recognize the specific change of hue of that dye.

HPLC and SERS revealed that sappanwood  is usually responsible for reddish brown, orange and pink shades (Catalog numbers 5, 9, 21), and Sophora for yellow shades in the designs (Catalog numbers 2, 5, 21). The two sometimes were used alone, sometimes mixed together. The field yellows are usually darker than the yellows in the designs (as in Catalog number 2) and often includes some sappan dye along with Sophora . The well known color fading/change of sappan is, therefore, probably the cause of the difference between the original and present hues. Moshe Tabibnia's analyses now provide scientific evidence for the hypothesis that there were Chinese carpets with reddish fields. Direct evidence is red field of the dragon chair cover and some antique specimens investigated for this exhibition showing reddish brown or brown-orange color at the knot base, where oxidation was minor.

Few exceptions to the traditional dyes were found: madder, Indian lake, an azoderivative red and alizarina for reddish hue, and  a dye allegedly containing gallotannins for some yellows.

The Indian lake was found in the field of the dragons chair cover (Catalog number 22): using HPLC, an unidentified molecule was found at the base of the knots, where an abnormal fading (pale yellow) appears with a spectrum close to yellow. The amount of that molecule is smaller in the yarns that are more red. The same lake red plus indigo blue are the dyes used for the purple tone; all of the blue in the designs is indigo.

Vis-RS allowed clarification of degrees of fading and color changes by recording the spectrum of the same dye at various parts of the pile (back, knot base, recto, tip). Fading is usually due to natural oxidation and exposure to light, seldom to mechanical wear. Yellow colors from Sophora are usually more intense at the knot base than at the tip of the pile. Some yellowish orange fields with no sappanwood (red) dye show no color change in their red component. As a rule, yellows in the fields of carpets are darker than those in the designs (Catalog number 21). Some yellow tones show a small amount of sappanwood, unrecognizable by Vis-RS but detectable by chemical analyses. Mustard yellow that is often found in the fields, as in a Ming item of XVIth century (Catalog number 3A), shows the spectrum of sappanwood and Sophora, but appears as a brown-orange hue at the base of the knots (Figure 69, inside of the pile of the Ming fragment). This because of the well known tendency of sappanwood to fade and change hue to brownish yellow upon long exposure to light. It is difficult to be sure of the original hue, but the knot base color is brown orange.

When the yellow has a stronger orange tone the presence of sappanwood is clear in the spectra. In other cases it is possible to see spectral bands typical of sappan/brazilwood in a fading orange that is more red at the knot than at surface, where are bands close to Sophora yellow. Vis-RS  shows that when sappanwood is mixed with a yellow dye there is less color change. Few exceptions to the traditional dye, madder (giving an orange hue), an azoderivative red (synthetic dye giving pink) and alizarina (synthetic dye giving orange) show spectra consistent with sappanwood. The synthetic dyes were used for repairing orange and pinkish yarns in  Ningxia in the second half of 19th century. The dragon chair cover (Catalog number 22) is an exception among antique Chinese carpet. It has a red field obtained from Indian lake. In fact, while some analytical tools report a mysterious pale yellow at the knot base and, therefore, a mysterious degradation of the red lake, Vis-RS does not detect any other dye. It gives an intense  Kerria lacca curve in the yarns on the back, where the red is better preserved, a little faded hue at the top of the pile and the same Kerria curve at the yellowish base. This indicates a chromatic change. Chinese dyers may have preferred the many refined red tones given by sappan/brazilwood until the 19th century, when synthetic colours arrived from West.

A yellow dye different from Sophora, a gallotannin source, was found in only one specimen, although quoted bibliography assigns safflower as the most common traditional yellow dye in the far east. Analyses on brown and black hues show gallotannins as main sources, confirmed by yarn corrosion. Logwood and Sophora are occasionally used for lighter tones. Vis-RS confirmed that purple was obtained from Indian lake plus indigo. Light and dark blue hues were all indigo, green hues were indigo plus Sophora.

There have been some problems identifying the sources of some hues in the analyzed carpets, a result of the fact that the dyes database is incomplete (2). Old dyers were really masters at exploiting natural sources of dyes. Were dyers aware of  fading and color changes with the dyes they used? In my opinion, the Tarim basin was definitively conquered by the Chinese emperor Qianlong in 1758, but commerce between China and this region from as early as the Shang dynasty (1600-1064 B.C.) is well documented. Jade from Khotan oasis was the usual trade item at that time. Chinese dyers knew how to get a lightfast red from madder, as we see in the Gansu and  Tarim basin masterpieces desplayed in the Moshe Tabibnia Gallery exhibition.

The catalogue numbers of the following images refer to the publication, M. Tabibnia, T. Marchesi (Eds), “Intrecci Cinesi”. Ancient Textile Art XV-XIX Century (Moshe Tabibnia: Milan, 2012)

Catalog number 2: Carpet with lotus blossoms, Pechino?, China, second half of XVI Century, 470 x 220 cm. Wool pile on a cotton foundation. (MATAM Collection)

Catalog number 3A: Carpet with dragons, Ningxia(?), China, XVI Century, 108 x 138 cm. Wool pile on a cotton foundation. (Gallery Moshe Tabibnia, Milan)

Figure 69: View of the pile yarn in Catalog number 3A.

Catalog number 5: Carpet with flowers, Pechino(?), China, second half of XVI Century, 325 x 225 cm fragment. Wool pile on a cotton foundation. (MATAM Collection)

Catalog number 9: Carpet with lotus flowers, Ningxia, China, mid XVII Century, 186 x 132 cm. Wool pile on a cotton foundation. (MATAM Collection)

Catalog number 21: Carpet with lotus flowers and holy mountain, Ningxia, China, ca. 1700 circa, 194 x 323 cm fragment. Wool pile on wool and cotton foundation. (Gallery Moshe Tabibnia, Milan)

Catalog number. 22: Chair cover with dragons, Ningxia, China, ca. 1700. Wool pile on wool and cotton foundation; 67 x 76 cm (Gallery Moshe Tabibnia, Milan)

Thank you for your kind attention,

Francesca Fiorentino


1. Silvia Bruni- Associate Professor of Analytical Chemistry, Department of  Inorganic, Metal-organic and Analytical Chemistry “Lamberto Malatesta”, Università degli Studi di Milano. Gianluca Poldi- Physicist, PhD Department of Arts, Fine Arts and Multi-media, Università degli Studi di Bergamo.
2. The database for Chinese dyes was set up with dyes drawn from the related vegetal species. Obviously, many of them are still unknown as gallotannins brown tones.



Home Page Discussion