Abstract
The application of natural dyes on natural fiber materials enables the realization of fully bio-based textile products. Usually natural dyes are applied in combination with mordanting agents to improve color strength and fastness properties. This current paper is dedicated to a comparison of the natural fibers wool and cotton with the regenerated cellulose fiber Lyocell. For natural dyeing, extracts from logwood or madder root are used as natural dyes. As mordanting agent, water soluble salts of Fe2+ and Cu2+ are evaluated in comparison to the less common agent titanium dioxalate, which is a water soluble chelate complex of Ti4+. Applications are done as pre- or meta-mordanting procedures. In this experimental set-up, 42 different sample combinations are realized. Depending on type of dye and mordant, deep color shapes are achieved in the range of red, black to blue. By application of titanium mordant color shades are realized which are not covered by dyeing procedures using the other both mordants. Compared to the dyeing without mordant agent, the lightfastness increases clearly with the metal mordant application. This statement is especially valid for logwood dyeing which exhibits low light fastness grades of 1 or 2 without metal mordant. Depending on mordant procedure, dye/mordant combination and fabric, a light fastness in the range of grade 1–5 can be realized. Best values are determined with dye/copper mordant combinations leading to a light fastness of grade 5, which is an excellent value for natural dyes. In comparison, the determined light fastness from titanium containing applications are minor. Compared to the dyeing without mordant agent, the rubbing fastness decreases clearly with the metal mordant application. However, the application with titanium mordant is advantageous in rubbing fastness and with the combination of titanium/madder, a wet rubbing fastness grade 4 can be reached, which is an excellent result for a natural dye application. The actual study presents a broad range of dyeing recipes for natural dyes. Especially reported is the natural dyeing of Lyocell fiber materials and titanium mordant agents, which in that combination unique and up to now less considered in literature. Finally, the actual study is a helpful tool and starting point for future developments of natural dyeing of regenerated fibers as Lyocell in combination with metal mordanting agents.
Keywords
Introduction
Fabrics made from natural fibers can be dyed by using several different types of dyes, as for example, acid dyes for wool fabrics or direct and reactive dyes for cotton.1 –4 Regenerated cellulosic fibers like viscose or Lyocell fibers are often dyed with reactive dyes.5,6 However, for dyeing of Lyocell fibers also procedures with vat dyes are reported.7,8 All these mentioned dyes are so-called synthetic dyes and are products of the chemical industry. Synthetic dyes have clearly several advantages as strong and reproducible coloration, good fastness properties and finally they are cost competitive.1,9,10 However, their bio-degradability is often low leading to difficulties during cleaning of dyeing waste water and the persistent distribution to the environment with the microfiber emission during customer washing.10,11 Alternatively, natural dyes can be used to avoid these disadvantages of synthetic dyes and offer a strategy for production of a fully bio-based textile product.12,13 An interesting approach in the field of natural dyes is their application using ionic liquids. One aim of this sustainable approach is the decrease in water consumption during dyeing processes. 14 Probable of the most innovative research papers in the field of natural dye on textiles is recently published by Eyupoglu et al. 15 These authors combine two innovative techniques – microwave-assisted heating and the prediction of dyeing recipes by using artificial neural networks. 15 To improve color intensity and color fastness, natural dyes are usually applied in combination with mordant agents, which are conventionally water-soluble metal salts from metals, for example, like aluminum, iron, copper, manganese, zinc, or tin.16 –22 Alternatively, bio-based textile chemicals are evaluated for mordanting procedures in respect to sustainability reasons.23 –26 Due to this, also for bio-based mordants, benefits and limitations are reported.27,28 In any case, there is a broad range of possible combinations of natural dyes with different metal mordant agents in different mordanting procedures. By these dyeing processes, a very broad range of different color shades and color strength can be realized even by using the same natural dye. Also, the fastness properties can vary in a broad range (from very weak to strong fastness grades).29,30 In contrast to the above-mentioned metal mordants and bio-based mordants, mordanting agents containing the metal titanium are less common, probable because there are less available water-soluble titanium compounds which can be used as mordant. Even the excellent and broad study of Feiz and Norouzi do not consider titanium mordants, even if these scientists consider nine different metal mordants for dyeing of wool with madder in three different mordanting procedures (pre, meta and post). 31 However, the use of titanium salts as mordants for silk dyeing is reported more than a century ago. 32 This early study by Hurst is not reporting on natural dyes. Instead synthetic dyes with alizarin basic structure are investigated for their interaction with titanium salts. 32 Two more recent studies on the use of titanium containing mordants are published by Cavaleri et al. and Yeo and Shin.33,34 The first study reports on wool dyeing with three different natural dyes applied in pre-mordanting procedure with four different metals (aluminum, iron, copper and titanium). The influence of the mordant on the color shade is here intensively discussed. However, fastness properties of dyed fabrics are less considered. 33 The second study is mainly dedicated to the natural dye gained from the red plant amaranth. Here the application is done on wool fabrics and the influence of three different metal mordanting agents (iron, zinc and titanium) in pre and post mordanting processes are investigated. 34 This study reports valuable data but is limited to only one less common natural dye extract. Another recent study by Zhou et al. reports on the use of titanium disulfate as mordant in combination with flavonoid dyes. 35 However, in that study the main investigation is focused on flame retardant properties and less on reached coloration and fastness properties. 35 By view on these interesting reports, it is clear that titanium mordants have some advantageous properties but are not fully investigated on different types of fabrics. Especially meta-mordanting procedures with titanium based mordants are less considered up to now. Based on this background, the actual study is focused on dyeing with two different natural dyes (extracts from logwood or madder root) with special focus on three issues – the comparison of pre-mordanting with meta-mordanting, the comparison of three different fiber types – including especially Lyocell and finally the comparison of titanium mordant to mordants based on iron and copper. Nevertheless, even if the use of metal mordants in textile dyeing is quite common, it should be mentioned that metal ions have also negative effects on human health, can lead to wastewater pollution and have potential ecotoxicological impact.36,37 The intensity of these health effects strongly depends on the type of metal and the solubility of the metal compound in water.38,39 Possible effects are for example, liver injury caused by metals. 40 However, as same as for every other chemical compound these effects are depending on the applied concentration. 41 For example, soluble iron compounds exhibit quite high LD50 values and are essential for human health if applied in lower concentration. 42 The LD50 values for iron salts are significantly higher as for copper salts, standing for a lower toxicity of iron ions compared to copper ions. However, it is remarkable that the limited concentrations for drinking water are lower for iron ions (0.2 mg/L) compared to copper ions (2 mg/L). 43 Titanium ions are not listed in European regulations on drinking water, because titanium is not related to negative health impacts. 43 Iron sulfate is even used in agriculture to improve plant growth. 44 In comparison to iron components, copper compounds exhibit a higher toxicity. 38 However, also copper components are used in different applications as antimicrobial agents. 45 Textile wastewater from dyeing or finishing processes using metal salts exhibits considerable amount of metal ion concentration. For environmental and health reasons these metal ions have to be removed from the textile wastewater. For this demand several removal technologies are available and used – prominent technologies are the based on flotation, coagulation, flocculation, and adsorption on different types of filter materials, as for example, activated carbon, zeolites or clay.46 –49 In conclusion, the use of metal components as mordanting agents should be carefully considered and done in respect to possible effects on human health and the environment. In the actual study, as dyed fiber materials additional to the natural fibers wool and cotton especially the regenerated fiber Lyocell is chosen, because natural dyeing systems on Lyocell are less reported up to now. The investigated fabrics differ in their chemical composition, because cotton and Lyocell fibers are built up by cellulose molecules and in contrast wool is built up by proteins. 50 For the cellulose based fibers, the metal ions from the mordants can only interact with the hydroxy groups on the fiber surfaces and by these complex bonds a fixation of the dye molecule on the fiber can be reached.51,52 In case of the protein based wool fiber several different interacting groups as the peptide unit, amino groups or carboxylic acids are available on the fiber surface to bond to the metal ions.53,54 By this, a more different dyeing behavior on wool can be expected in comparison to the other two investigated fibers cotton and Lyocell. However, even if both types of fibers cotton and Lyocell are based on molecular level on the biopolymer cellulose, different dyeing behavior can be expected. Dyeing properties on these fiber materials can differ, because of different crystallinity of cellulose and different porosity of the fiber surface.55,56 Further, a different capability for the up-take of metal ions as part of mordanting agents is reported for cotton and Lyocell fibers. 57 At nearly neutral pH of 8, cotton fibers exhibit a better capability for the up-take of iron ions. In contrast, under strong alkaline conditions Lyocell fibers show a better capability for bonding of iron ions. 57 For this, dyeing of Lyocell fibers is a special focus of the actual study, if natural dyeing recipes can be successfully transferred to regenerated cellulose fiber materials. By the actually presented broad experimental set-up, 42 different dyeing preparations are realized. Following this, a complete view on the influence of different metal mordant agents and mordanting procedures is supported for reached coloration and fastness properties. Depending on type of dye and mordant, different color shades are gained in the range of red, black to blue. By application of the especially considered titanium mordant, color shades are realized which are not covered by dyeing procedures using the other both investigated metal mordants. In comparison to the dyeing results realized without mordant agent, the lightfastness increases clearly in presence of metal mordant. This observation is especially valid for dyeing with logwood leading to only low light fastness grades of 1 or 2 without metal mordant. Depending on used mordant procedure, dye/mordant combination and type of fabric, a light fastness in the range of grade 1 to 5 can be realized. Best values are gained with dye/copper mordant combinations leading to a light fastness of grade 5, which is an excellent grade for natural dyes. In contrast, the light fastness samples dyed with titanium mordant are minor. Compared to the dyeing without mordant agent, the rubbing fastness decreases clearly with the metal mordant application. However, the application with titanium mordant is advantageous in rubbing fastness. With the combination of titanium/madder, a wet rubbing fastness grade 4 can be reached, which is a good grade for application of natural dyes. The actual study presents a broad range of dyeing recipes for natural dyes. Especially considered is the natural dyeing of Lyocell fiber materials and titanium mordant agents, which is in that combination unique and up to now less considered in literature. Finally, the actual study is a helpful tool and starting point for future developments of natural dyeing of regenerated fibers as Lyocell in combination with metal mordanting agents.
Experimental section
Materials
Dyeing experiments are performed on three different types of fabrics made from wool, cotton or Lyocell fibers. The wool and cotton materials are plain weaved fabrics with weight/area of 230 and 180 g/m2, respectively. The Lyocell material is a knitted fabric (single Jersey) with weight/area of 180 g/m2. Microscopic images illustrating structure and topography of the used fabrics are presented in Figure 1.

Microscopic images of the used fabrics. The images are recorded by scanning electron microscopy (SEM).
Two different types of natural dye stuffs are used, an extract from madder root (Rubia tinctorum) and an extract from logwood (Haematoxylum campechianum). Both dye extracts are supplied as water soluble powders and commercially available products supplied by the company Pflanzenfärbershop (Hückelhoven-Baal, Germany) together with basic information and suggestion and dyeing recipes.58 –61 Recently infrared spectroscopic measurements on both used dye extracts are published and discussed in comparison to other natural dye products.62,63 The color components in extract from madder root are reported to be a mixture from several different anthraquinone components, with the main color component alizarin (compare chemical structure in Figure 2).64,65 Other researchers identify and investigate the anthraquinone derivative purpurin as second key color component of madder root. 66 The main color component in logwood extract is hematein (Figure 2).64,67 In interaction with metal mordant, different types of coordination of metal ions to the molecules of alizarin and hematein are presented in literature.68 –70 For alizarin, two types of coordination are reported. First, the metal ion is coordinated by the carbonyl group and the neighbored hydroxy group. Second, the metal ion is coordinated between the two hydroxy groups. 68 For the hematein molecules also different types of possible coordination of metal ions are proposed. There is the coordination of one metal ion to two neighboring functional groups – carbonyl and hydroxy or between two hydroxy groups.69,70 Further, the coordination of the metal ion to only one functional group of the dye molecule is proposed. 70 Puchtler et al. also mention the formation of polynuclear complexes, where several dye molecules of hematein are bonded together by several metal ions. 70

Chemical structures of the main color components in madder root and logwood extracts – (a) Alizarin and (b) Hematein.
The dyeing is done with combination of three different mordanting agents, iron(II)sulfate heptahydrate FeSO4 • 7H2O, copper(II)sulfate pentahydrate CuSO4 • 5 H2O, and titanium(IV)dioxalate Ti(C2O4)2. The iron(II)sulfate is gained from Merck KGaA (Darmstadt, Germany), the copper(II)sulfate is gained from Carl Roth GmbH (Karlsruhe, Germany) and the titanium(IV)dioxalate is supplied by Jeromin Shop (Mannheim, Germany). The titanium(IV)dioxalate is a water soluble chelate complex (compare chemical structure in Figure 3). For washing and cleaning of the fabrics the washing agent Perlavin NIC from Textilchemie Dr. Petry GmbH (Reutlingen, Germany) is used. 71

Chemical structure of titanium (IV) dioxalate – Ti(C2O4)2 (Mw = 223.9 g/mol).
Dyeing procedures
Initially before dyeing experiments all used fabrics are washed with an aqueous solution of 2 g/L Perlavin NIC at temperature of 70°C. After washing the fabrics are rinsed twice with soft water of 60°C temperature and again twice washing is done with soft water of 30°C temperature. Washing and rinsing procedures exhibit a duration of around 20 min. After washing the fabrics are line dried at room temperature. Finally, the dried fabrics are cut into pieces containing a weight of 10 g each, which are used for the further dyeing experiments. All dyeing and mordanting procedures are done with the dyeing machine Ahiba Polymat (Datacolor GmbH, Marl, Germany) in a ratio of 10:1 (100 mL liquid and 10 g fabric). The temperature regime is set for all applications similar with the process temperature of 80°C kept for 60 min duration. The heating rate is set with 2°C/min. After application the liquid recipe is cooled down to 50°C and following the fabric is removed from the liquid. The total process duration is around 90 min. For the main experiments, the dye extracts are applied together with the mordanting agents. To support also reference materials for reference color determinations, two types of reference treatments are done (compare Figure 4). During these reference treatments either the mordanting agent or the dye extract are applied alone without the other agent. To produce reference samples with mordanting agent, aqueous solution of 0.3 wt-% of the metal salts are used. For production of reference samples with dye, a solution of the dye extracts with a concentration of 5%-ofw is applied. For application of dye with mordanting agents two different procedures are used, pre-mordanting and meta-mordanting (compare Figure 4). For the pre-mordanting procedure, in a first step the mordanting agent is applied in an aqueous solution with 3%-ofw and in a second step the dye extract is applied with 5%-ofw. For the meta-mordanting procedure, an aqueous solution containing 3%-ofw mordanting agent and 5%-ofw dye extract is applied in one single step on the fabrics. The applied concentrations for dye extract and mordanting agents are chosen based on two previous studies related to dyeing with natural wood extracts under usage of aluminum and iron metal mordants.30,72 These studies support a comparison on different mordanting procedures (pre-, meta- and post-mordanting) with different concentration of dye extract and iron-mordant on cotton. Earlier it was shown that for the current system post-mordanting lead to inferior results, so it is not considered now.30,72

Schematic drawings of the different used procedures – reference treatments, pre-mordanting and meta-mordanting.
Analytics
The coloration of the prepared samples is determined by measurement of the CIE-Lab indices and the color strength (K/S) using a device Datacolor SpectraVision (Datacolor GmbH, Marl Germany). With this device additionally photographs of the samples are made to support a visual color impression. The evaluation and discussion of determined CIE-Lab indices and their relation to color depth and shade is done according to standard literature. 73 The rubbing fastness of dyed samples is determined following the standard DIN EN ISO 105-X12 by using a crockmeter (James Heal, Sterling, Virginia, USA) as rubbing device. Determined are the dry and wet rubbing fastness by rubbing a dry or wet cotton test fabric on the dyed sample. The determination of the grade of rubbing fastness is done by comparison of the rubbed test fabric to a gray standard (according to standard ISO 105-A03). The scale of rubbing fastness is in between grade 1 (lowest rubbing fastness) to grade 5 (excellent rubbing fastness). The light fastness of dyed fabrics is determined according to standard ISO 105-A02 using the testing device Xenotest Alpha LM instrument (Atlas Material Testing Technology GmbH, Linsgengericht-Altenhaßlau, Germany). During illumination the relative humidity is set to around 40%. All samples are illumination two times for 55 h. Accompanying a blue standard is illuminated and by comparison to this illuminated blue standard the grade of light fastness is determined. The scale of light fastness is in between grade 1 (lowest light fastness) to grade 5 (best light fastness). The illumination of samples is done until the grade 5 is reached for the blue standard. For recording microscopic images, the scanning electron microscopy (SEM) is applied. Here a microscope TM4000 Tabletop (Hitachi, Japan) is used. Infrared spectra (IR-spectra) of dyed textile samples are recorded using a FT-IR spectrometer IRTracer-100 from Shimadzu (Japan) equipped a Specac Golden Gate ATR unit. For each sample measurement, the number of scans is set to 20.
Results and discussion
Reference materials
The reference materials are the untreated fabrics and the fabrics which are only treated with the mordant agent without dye. The Table 1 illustrates the coloration by showing the sample photographs together with the determined CIE Lab indices. The untreated fabrics show a white appearance, documented with high L* indices in the range of 82–90. Most bright and white reference is the cotton fabric with L* = 90 and the indices a* and b* near zero. In comparison, the wool and Lyocell fabrics exhibit especially for index b* higher values of 3.6 to 8.1 standing for a certain yellow shade of coloration of these fabrics.
Coloration of reference fabrics without application of natural dyes.
Shown are the photographs of the samples and related CIE Lab indices.
The application of the iron mordant clearly changes the fabric coloration to orange/brown. The index L* is clearly decreased. The indices a* and b* are significantly shifted to positive values standing for a shift to red and yellow (orange coloration). The strongest color change is determined for the wool fabric and the lowest one for the cotton fabric. The orange coloration is probable the result of an oxidation of the applied Fe2+ ions to Fe3+ containing oxide species, due to typical yellow/red coloration of iron oxide materials.64,74
The application of copper and titanium on cotton and Lyocell fabrics leads only to slightly shift in coloration. Here only a slight shift to green can be determined. In comparison, for the wool fabric a stronger color shift can be determined. For copper on wool a darker green can be determined reflected by the clearly negative value of index a* = −9.0. This is the most negative value for a* for all here compared samples. The application of the titanium mordant on wool fabric leads to a light shift to yellow coloration. However, compared to the other mordanting agents on wool, with titanium mordant the color change caused by the mordant itself is the lowest.
Coloration results
Application on wool fabrics
The coloration of dyed wool fabrics is documented as photographic images together with the CIE lab indices and the K/S values in Table 2. To illustrate the color shifts according to the indices a* and b*, these indices are also shown for prepared samples in Figure 5. The logwood dyed wool fabric without any mordant application exhibits a dark brown coloration with quite high positive values for the indices a* and b*. By combination with mordants, the color shade is clearly different. With the addition of the mordants iron and copper, a coloration in the range of black to blueish black is reached for the logwood samples. Here with meta-mordanting, the realized samples are a bit lighter and more blueish (more negative b* indices). In contrast, by application with titanium mordant completely different color shades are reached, which could be described as orange brown and yellowish black. With pre-mordanting, both indices a* and b* are significantly positive, standing for the orange brown color shade. For the meta-mordanting, a clearly different result is gained with a lighter sample and yellow color shade. The color strength (K/S) strongly depends on the type of mordant and pre-mordanting procedure. The application of copper mordant in a pre-mordanting procedure leads to a significantly increased color strength compared to the dyeing process without mordanting agent. With iron or titanium mordants in pre-mordanting, nearly the same or slightly stronger color strength is determined. In contrast, with meta-mordanting procedures in this investigated logwood application the color strength is decreased compared to the dyeing process without mordanting agent.
Coloration of dyed wool fabrics.
Shown are the photographs of the samples, related CIE Lab indices and the K/S values.

Color variation of prepared wool fabrics dyed – (a) logwood/(b) madder. Shown is the relation of the two indices a* and b* for the different mordants (Fe, Cu, Ti) in different application methods (pre/p and meta/m). The not mordanted sample is signed with d.
All madder dyed wool fabrics exhibit a deep red coloration. However, the intensity of red and the color shade of the red is different depending on mordant agent and mordant procedure. The madder application on wool without mordant leads to a clear red color shade with a certain content of yellow. Only with application of iron or titanium mordant in the pre-mordanting procedure, the mordanting samples exhibit a darker shade, as reported earlier by Feiz and Norouzi for iron mordanting with madder. 31 Here an increased color strength (K/S value) is determined compared to K/S of the dyed wool fabric without mordanting agent. All meta-mordanting samples exhibit a decreased darker coloration compared to the sample without mordant. The sample (pre-mordanting/iron) is best described with darker and less yellowish red color shade. The samples with copper mordant (both pre- and meta mordanting) exhibit nearly the same color strength and the red shape is less yellowish. For this, if a pure and deep red coloration is demanded, these three combinations are advantageous. In contrast the application of the iron mordant in meta-mordanting process leads to a strong yellow color shift. Finally, the coloration results for the samples with titanium mordant are completely different. Here highest values for both indices a* and b* are determined standing for an orange to orange/brown coloration, which is clearly different from the typical madder red.
Application on cotton fabrics
The coloration results on cotton fabrics are documented in Table 3 as photographs together with the CIE Lab indices and K/S values for each sample. To support an enhanced view on the color shifts caused by the mordanting processes, the indexes a* and b* are compared for the different samples in two further graphs (Figure 6). The logwood application without mordant leads to a red/brown coloration with a quite high index a* of 15.9. The combination with the mordant agents leads in all cases to a significant color shift and in most cases to a stronger color depth and increased color strength (K/S). With iron mordant, the color depth increased significantly, documented by clearly lower indices L* 20 (pre-mordanting) and 24 (meta-mordanting). Here, the K/S values increase to more than 6 compared to K/S = 2.0 for the cotton sample dyed without mordant. For these logwood/iron samples, the color appearance can be best described as blueish black, which is accompanied by clear negative values for the index b*. Here with meta-mordanting procedure, the index a* reaches a value of 3, indicating a slightly red-shift of the blue coloration to a finally blacker/bluer/violet appearance. By application of the logwood/copper modifications on cotton also a blueish black color shade is achieved. However, the blue shift is even stronger with more negative values for index b* – most negative value for meta-mordanting procedure with b* −11.2. As the mordanting with iron and copper leads with logwood to quite similar color shade, in comparison the mordanting procedure with titanium gives clearly different results. Also, the difference between pre-mordanting and meta-mordanting is with titanium mordant agent quite strong. The combination logwood/titanium applied in pre-mordanting leads nearly to the same brightness as the dyed sample without mordant. For this combination, even the color strength (K/S) is nearly similar. However, the index a* is smaller and index b* is negative. This coloration can be described with dark violet/red. In contrast, the same combination logwood/titanium applied in meta-mordanting leads to darker coloration without any red component in color shade. This coloration can be described as black with blueish/greenish component. Finally, it is seen that the color shade with titanium mordant is different to the other mordants (iron or copper) and titanium mordant is also more sensitive to the type of mordanting procedure (pre or meta).
Coloration of dyed cotton fabrics. Shown are the photographs of the samples, related CIE Lab indices and the K/S values.

Color variation of prepared cotton fabrics dyed – (a) logwood/(b) madder. Shown is the relation of the two indices a* and b* for the different mordants (Fe, Cu, Ti) in different application methods (pre/p and meta/m). The not mordanted sample is signed with d.
The application of madder on cotton without mordant leads to light red coloration, which can be described as strong rose. In comparison, the combination of madder/iron leads to darker coloration with a clearly decreased index L* and increased color strength (K/S). However, also the red color shade is decreased with significantly decreased index a*. In case of meta-mordanting with madder/iron also yellow color component with index b* 6.5 is observed. Here, the iron mordant leads to higher K/S value but also to a decrease of the typical madder red color shade. A different coloration result is gained for the combination of madder/copper. Applied with pre-mordanting procedure using copper mordant, nearly the same color shade as the not mordanted sample is gained but with increased color strength – shown also by the clearly decreased index L* and the increased K/S value. The achieved color can be described as dark rose. The dyeing application madder/copper in meta-mordanting leads in comparison to lower color strength, which is in between the color strength of the not mordanted and the pre-mordanting result. The result with the combination of madder/titanium is clearly different and can be described as strong red coloration. Here, the index L* is clearly decreased and the color strength (K/S) is strongly increased. The index a* is around the double of a* for the not mordanted application. If a strong red coloration without a rose component is wished as aim of the dyeing process, probable the titanium mordant is advantageous compared to the mordanting with iron or copper.
Application on Lyocell fabrics
Compared to the application on wool or cotton, the reached color strength is on Lyocell fabrics lower. The lower color strength reached on Lyocell can be explained mainly by lower dye up-take of the fiber and also by different fiber dye interaction. Cotton and Lyocell fibers are both built up by cellulose and should exhibit quite similar fiber-dye-mordant interaction. Nevertheless, due to differences in porosity and crystallinity also for these two fibers different dye up-take and realized color strength can be expected.55,56 Nevertheless, also on Lyocell fabrics intensive coloration can be reached by using the actually presented dyeing recipes. The reached coloration on Lyocell fabrics are presented in Table 4 for each prepared sample together with its CIE Lab indices and the K/S values. To improve the discussion on the color shift, the indices a* and b* are presented and compared separately in two graphs (Figure 7). Applied on Lyocell the logwood without mordanting leads to coloration best described as dark/rose or reddish black. The index a* is with 20.9 quite high standing for a clear red color component. This coloration result is clearly different from the dyeing of wool and cotton with the same recipe, which lead to a brown/red coloration. On Lyocell fabric, also the combination of logwood with iron or copper mordant leads to stronger color depth and clear change to different color shades which can be described as black or blueish/black. For these dye/mordant combinations, the color strength (K/S) is clearly increased. With logwood/iron combination in pre-mordanting, a black to dark-gray coloration can be realized. The same combination of logwood and titanium mordant in meta-mordanting results in a dark-gray with blueish component and the highest color strength (K/S) realized in the current study on Lyocell fabric. The Lyocell samples with the combination of logwood/copper exhibit a blueish black color shade. Completely different coloration results are gained with titanium mordant. The application of the combination logwood/titanium by pre-mordanting procedure leads to a quite similar coloration result and a slightly decreased color strength (K/S) as gained with the logwood applied without mordanting agent. In contrast, the same combination logwood/titanium applied in meta-mordanting procedure leads to completely different coloration result with a color described as greenish black. Here the color strength (K/S) is significantly increased. Also for logwood dyeing on Lyocell, the coloration with titanium mordant is strongly sensitive to the type of mordanting procedure pre or meta.
Coloration of dyed Lyocell fabrics.
Shown are the photographs of the samples, the related CIE Lab indices and the K/S values.

Color variation of prepared Lyocell fabrics dyed with logwood – (a) logwood/(b) madder. Shown is the relation of the two indices a* and b* for the different mordants (Fe, Cu, Ti) in different application methods (pre/p and meta/m). The not mordanted sample is signed with d.
The application of madder on Lyocell fabric leads to a rose color shade with only medium color strength. With all three metal mordants the color strength (K/S) is increased and a change in color shade is caused. The differences in color shade according to the different metal mordants are significant. In comparison, the effect of the type of mordanting procedure (pre or meta) is less. With madder/iron combination a reddish gray color shade is gained which is not comparable to the typical madder red coloration. In comparison, with the combination of madder/copper a coloration dark/rose can be reached, which exhibits a color shade quite similar to the not mordant sample but with stronger coloration intensity. Finally, with the combination of madder/titanium a coloration is reached which can be clearly described as full red – shown also in an index a* with is around double of the a* value of the not mordanting madder Lyocell sample. If a red coloration on Lyocell fabric is demanded under the investigated recipe combinations, the titanium mordant with madder is advantageous. In case a rose coloration is aimed instead of titanium better a copper based mordant agent is recommended to use.
Fastness properties
Light fastness
The light fastness of all dyed samples prepared without mordanting agents are given in Figure 8. The logwood dyed fabrics exhibit very weak light fastness with grade 1 and 2. For madder dyed fabrics better results with grade 3 to 3.5 are reached. The anthraquinone based chromophore of the madder color components exhibits obviously a better light stability compared to the hematein in the logwood extract. Compared to the dyed wool and cotton fabrics, the dyed Lyocell fabric shows slightly better light fastness. This result is a bit uncommon, because the coloration intensity of the dyed Lyocell fabrics is lower and usually a lower coloration intensity is accompanied with lower light fastness.

Light fastness for prepared fabrics dyed without mordanting agent.
The light fastness of all dyed wool fabrics with applied mordanting agents are presented in Figure 9. For application of logwood, in most cases the light fastness is significantly improved by application of the different mordanting agents. For iron and copper mordant applied in pre-mordanting, light fastness grades of 4 or 5 are determined. In comparison, with meta-mordanting the light fastness is around one grade number lower but still on a good level. A light fastness of grade 5 is for dyeing with natural dyes an excellent value. For dyeing with the combination logwood/titanium in pre-mordanting only very weak light fastness is observed. However, with meta-mordanting procedure this combination leads to a better result with grade 3.5. For dyeing with madder, the results are different. Because madder dyeing without mordanting on wool samples already shows a good light fastness grade 3, the wool samples dyed with madder mordant combinations exhibit same or only slightly improved light fastness. Best light fastness with madder on wool is gained for the combination of madder/iron applied in meta-mordanting process with a grade of 4.5, which is slightly below the best light fastness with the logwood/copper combination.

Light fastness for wool fabrics dyed with mordanting.
The light fastness grades for all dyed cotton samples with different mordants are summarized in Figure 10. Compared to the results on wool, for the dyed cotton fabrics the gained light fastness is stronger determined by the type of used mordant agent. A rough ranking can be set with copper > iron > titanium, with copper mordant leading to best light fastness grades. The advantage of copper mordant to reach higher light fastness for madder dyeing on cotton is also reported in literature for the comparison of copper and aluminum mordant. 75 For the iron mordanting with logwood or madder on cotton, all combinations lead nearly to similar light fastness around grade 3. In comparison to the logwood application without mordant, this is a clear improvement in light fastness. The determined light fastness of cotton fabrics dyed with logwood/iron combination is slightly better compared to lower light fastness grades reported in literature. 51 However, for madder the combination with iron mordant does not lead to any further improvement of the light fastness. In contrast, the copper mordanting leads to significant improvement of light fastness, both in combination with logwood or madder. With madder/copper recipes on cotton even high light fastness with grade 5 is determined, which is an excellent value for a natural dyeing system. The application of titanium mordant, does not lead to any improvement of light fastness or even in case of the combination madder/titanium to a decreased light fastness compared to the sample dyed without mordanting agent. By view on this weak light fastness results, titanium mordants are clearly disadvantageous for applications where a good light fastness is required.

Light fastness for cotton fabrics dyed with mordanting.
For dyed Lyocell fabrics with mordant applications, the light fastness values are compared in Figure 11. The results are nearly comparable to the light fastness determined for similarly dyed cotton fabrics. Recipes with copper mordant lead to good light fastness. With the combination logwood/iron a certain improvement of light fastness is reached. However, the light fastness of madder is nearly not improved by its combination with iron mordant. With titanium mordant, in some combinations the light fastness of dyed Lyocell samples is even decreased.

Light fastness for Lyocell fabrics dyed with mordanting.
A lower light stability in contact with titanium compounds might be explained with photocatalytic processes which can lead to photocatalytic decomposition of dye stuffs.76,77 However, the photocatalytic property is related to the crystalline type anatase of titanium dioxide. Other types of titanium dioxide (like rutile) or amorphous components are not photoactive.78,79 The formation of anatase is probable unlikely under the low process temperatures, so a photocatalytic property is probable not the reason for the lower light fastness. 79 The lower light fastness of samples with titanium mordant is probable simply a result of the lower light stability of the dye titanium complex compared to higher light stability of dye iron and dye copper complex. For comparison, the earlier study of Zhou et al. comparing iron and titanium based mordant in combination with flavonoid dyes does not show a significant decrease in light fastness for the titanium mordant. 35 For this, the actually reported decrease in light fastness might be specifically related to the combination of titanium mordant with logwood and madder and cannot be generalized to other types of natural dyes.
Rubbing fastness
The rubbing fastness of dyed samples without mordanting agents are in the range of 3 to 4.5 (Figure 12). Best values are gained for the application with madder tested for dry rubbing fastness. With wet rubbing slightly lower or the same rubbing fastness is reached. Such a wet rubbing fastness grade of 4 can be seen as a good rubbing fastness especially for natural dye application. Higher wet rubbing fastness would be expected for reactive dye application. The rubbing fastness of logwood dyed samples is for dry rubbing test nearly on the same level as the results for the madder dyed samples. However, with wet rubbing the logwood dyed samples exhibit clearly lower performance with a grade of only 3. Probable the adhesion of the madder components to the fiber surface is advantageous compared to the adhesion of the logwood components. In the actual comparison, the type of fabric – wool, cotton or Lyocell – has nearly no influence on the determined rubbing fastness of dyed fabrics.

Rubbing fastness for prepared fabrics dyed without mordanting agent.
The rubbing fastness determined on dyed wool fabrics with different mordants are presented in Figure 13. The left image is dedicated to logwood applications and the right image to dyeing with madder. All wool samples dyed with mordant containing recipes exhibit lower rubbing fastness, this can be explained by the increased color strength of these samples caused by the mordant. Dyed samples with stronger coloration are more likely to exhibit lower rubbing fastness. However, the intensity of this decrease in rubbing fastness strongly depends on the type of applied natural dye and its combination with the mordant. The combination of logwood with iron or copper leads only to weak rubbing fastness grade of 1–2. These combinations are mostly not suitable for textile application, by view on the rubbing fastness. Surprisingly, the combination of logwood/titanium leads to quite good rubbing fastness. For this combination a grade of 3 is reached even for wet rubbing. Obviously, the applied titanium ions initiate a better adhesion of the logwood dye to the wool fiber as iron or copper ions can do. One possible explanation for this better adhesion might be the stronger positive charge of the Ti4+ ions compared to the Cu2+ and Fe2+ ions from the other mordants. For the madder applications on wool, the influence of the different mordant agents on the rubbing fastness is less strong. Here, for most recipe combinations, a medium rubbing fastness grade 3 is reached. For iron mordant, the dry rubbing fastness is determined with grade 3, which is better as values recently reported in literature for a printing application of madder on wool fabric. 80 In actual investigation, best rubbing fastness on wool gained with madder is determined in combination with the titanium. However, only if this is done in pre-mordanting procedure. For that case, even for wet rubbing a rubbing fastness grade 4 can be reached, which is same as for the not mordant dyed sample and is a good value for a natural dye application.

Rubbing fastness for wool fabrics dyed with mordanting. (left image: logwood/right image: madder). Rubbing fastness for wool fabrics dyed with mordanting – (a) logwood/(b) madder.
For dyed cotton fabrics with mordanting, the rubbing fastness is presented in Figure 14. Separately shown are the rubbing fastness for logwood dyed cotton samples (left image) and madder dyed cotton samples (right image). With logwood application on cotton significantly lower rubbing fastness can be achieved as for applications with madder. The combination of logwood with the mordants iron or copper leads only to weak rubbing fastness of grade 1–2. The combination of logwood/titanium leads on cotton fabrics only to a slight improvement of rubbing fastness. For the madder dyed cotton fabrics, the rubbing fastness is in the range of 2.5–4 and by this clearly better compared to the logwood dyed samples. Similar to wool, also on cotton the combination madder/titanium applied in pre-mordanting procedure leads to best rubbing fastness among the other cotton samples. Nevertheless, here also with the combination madder/copper good dry rubbing fastness of grade 4 and an intermediate wet rubbing fastness grade 3 are reached. This issue is especially valuable, because with the combination madder/copper in meta-mordanting also high light fastness grade 5 is determined. With this preparation a good combination of different fastness properties is reached.

Rubbing fastness for cotton fabrics dyed with mordanting. (left image: logwood/right image: madder). Rubbing fastness for wool fabrics dyed with mordanting : (a) logwood/(b) madder.
The rubbing fastness results for dyed Lyocell fabrics with mordanting are presented in two images in Figure 15. The logwood dyed Lyocell fabrics exhibit a moderate low rubbing fastness of around 2. The dry rubbing fastness for these samples is slightly stronger in case of copper and titanium mordanting. Here, the mordant application clearly decreases the rubbing fastness compared to the logwood dyed sample without mordanting. In comparison the madder dyed Lyocell fabrics show a better performance in the rubbing test. Most of these samples exhibit a rubbing fastness around grade 3, even in case of wet rubbing. With the combination of madder/titanium applied in pre-mordanting procedure best wet rubbing fastness grade 4 is determined, which is similar to the performance of the madder dyed Lyocell fabrics without mordanting agent. This is an especially good result, because this Lyocell fabric with madder/titanium shows a deeper coloration compared to the madder dyed Lyocell fabric without mordanting application.

Rubbing fastness for Lyocell fabrics dyed with mordanting. (left image: logwood/right image: madder). Rubbing fastness for wool fabrics dyed with mordanting: (a) logwood/(b) madder.
IR spectroscopic investigations
IR spectroscopic is an analytical method to identify fiber materials. 46 IR spectroscopy can be also used to investigate dye stuffs or natural dye extracts.62,63 For these reasons, the IR spectroscopy is used to analyze the actually prepared colored textiles sample with the aim to determine the applied dye onto the fiber material. However, it was reported earlier that the vibrational signal in the IR spectra cause by the fiber are strongly dominate and the dye molecules cannot be identified by IR spectroscopy on dyed fabrics. Only in case of deeply colored fabrics, few weak signals can be assigned to the dye molecules. 46
The Figures 16 to 18 present the IR spectra of actually prepared colored fabrics. Reference IR spectra are recorded from the undyed fabrics. For all three types of fabrics – wool, cotton and Lyocell – the typical IR spectrum is determined and the fiber material can be clearly identified by comparison with IR spectra from literature.46,81 The IR spectra recorded from different cotton and Lyocell fabrics exhibits nearly the same pattern (Figures 17 and 18). No additional peak appears which could be related to the presence of the applied dye molecules. For these fiber types, on current samples it is not possible to determine the dye molecules on the fiber surface. A slightly different result is gained for the IR spectroscopic investigation of the dyed wool samples. The IR spectra of the wool samples dyed without mordant or with iron or copper mordant exhibit the same pattern in the IR spectrum as the undyed wool reference (Figure 16). For these dyed samples, no additional peak appears which could be related to the presence of the applied dye molecules. Also, on the wool fabric the dye molecules cannot be identified by IR spectroscopy. However, all four wool samples dyed with the titanium mordant exhibit a clear additional signal at 1317 cm−1. A signal in this spectral region probable not related to any type of vibration related to Ti-O-C unit which could be part of a fiber metal complex. Vibrations related to Ti-O-C or Ti-O-Ti units appear in the spectral region below 1180 cm−1. 82 It is reported in literature that the denaturation of proteins can influence their IR-spectra, also in the spectral region around 1300 cm−1.83,84 Considering a partly denaturation of the wool protein in presence of the Ti4+, the additional signal at 1317 cm−1 can be explained for the wool samples dyed under titanium mordanting.

IR spectra recorded from differently dyed wool fabrics – image (a) for dyeing with madder and image (b) for dyeing with logwood.

IR spectra recorded from differently dyed cotton fabrics, dyeing process with meta-mordanting procedure – image (a) for dyeing with madder and image (b) for dyeing with logwood.

IR spectra recorded from differently dyed Lyocell fabrics, dyeing process with meta-mordanting procedure – image (a) for dyeing with madder and image (b) for dyeing with logwood.
Summary and conclusion
The current study supports a broad experimental set-up to evaluate realizable color shades and fastness properties on different types of textile materials. Three types of textile materials are used – fabrics from wool, cotton and Lyocell fibers. Two types of natural dyes are used – extracts from logwood and madder root. Three types of metal mordanting agents are used – iron, copper and titanium. Two types of different mordanting procedures are used – pre-mordanting and meta-mordanting. By this, in summary 42 different textile sample combinations are realized. This broad set-up offers a detailed view on the influence of each parameter on color shade and fastness properties. In contrast, to many earlier publications this actual report considers especially Lyocell fabrics in comparison to natural fibers, which is seldom reported for natural dye application. Also, the use of titanium metal mordant is seldom reported in literature. For this, the current experimental set-up offers new and detailed results for natural dyeing applications. Especially the reported application of natural dyes in combination with titanium based metal mordants on Lyocell fabrics is of high novelty. Reference dyeing experiments are done without metal mordant. Here, the reached color strength is only moderate but the rubbing fastness is good. The light fastness is mainly determined by the type of dye, probable depending on the light stability of the chromophore. By combination with the metal mordants, deep color shades can be realized in the range of red, black to blue. These color shades are strongly depending on all four parameters – type of textile, type of dye, type of mordant, and the kind of mordanting procedure. The reached color strength decreases from wool to cotton to Lyocell fabrics. However, even on Lyocell fabrics intensive colorations can be realized. Upon the mordant agents, the application of titanium mordant often leads to a special color shade quite different from the results gained with iron or copper mordants. Compared to the dyeing without mordant agent, the lightfastness increases significantly with the metal mordant application. This observation is especially made for samples from logwood dyeing exhibiting only low light fastness grade of 1 or 2 without metal mordant. Depending on mordant procedure, dye/mordant combination and fabric, a light fastness of grade 1–5 can be observed. Best values are gained with dye/copper mordant combinations leading to a light fastness up to grade 5, which is a good result for natural dyes. In comparison, light fastness of dyed samples with titanium mordant are minor. Compared to the dyeing without metal mordant, the rubbing fastness decreases clearly with the metal mordant application. However, dyed samples with titanium mordant are advantageous in rubbing fastness. With the combination of titanium/madder, wet rubbing fastness grade 4 can be reached, which is a good result for a natural dye application. Finally, the actual study is a helpful tool and starting point for future developments of natural dyeing of regenerated fibers as Lyocell in combination with metal mordanting agents. Such future developments should overcome the limitations of the actual study and include also the up-scale of the dyeing process, repeating trials on larger scale and cost-calculations. These next steps are crucial for commercial and industrial applications but going beyond the scope of the actual study which is mainly supporting a proof-of-concept and starting point for further developments.
Footnotes
Acknowledgements
The authors owe many thanks to Dipl.-Ing. Simone Wagner for help during dyeing experiments in the finishing lab of the Hochschule Niederrhein. For assistance during de-termination of light fastness grades the author owe many thanks to Mrs. Petra Lünskens and Mrs. Afsana Akhter. All product and company names mentioned in this article may be trademarks of their respected owners, even without labeling.
Author contributions
Sample preparation, color measurements and rubbing fastness testing, AnikaOstheim; preparation of Lyocell knitted fabrics and writing, Karin Ratovo, Kristina Klinkhammer & Madita Fischer; organization of light fastness tests, Karin Ratovo; supervision & resources, Ellen Bendt; conceptualization, writing, original draft preparation, review and editing, visualization & supervision, Boris Mahltig. All authors have read and agreed to the published version of the manuscript.
Funding
The authors disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: Funded by the European Union. Views and opinions expressed are however those of the authors only and do not necessarily reflect those the European Union or the European Research Executive Agency (REA). Neither the European Union nor the granting authority can hold responsible for them. Grant agreement ID: 101134042.
Declaration of conflicting interests
The authors declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.
