“Purplish Blue” or “Greenish Grey”? Indigo Qualities and Extraction Yields from Six Species

Indigo quality is determined by its indigotin content. Another quality indicator is colour. For an evaluation of species, indigo samples from Indigofera tinctoria, Indigofera suffruticosa, Indigofera arrecta, Persicaria tinctoria, Strobilanthes cusia and Wrightia laevis cultivated in Austria and China were visually classified and analysed spectrophotometrically and using a L*a*b* measuring device. In addition to a standardised hot-extraction method without lime, some samples were extracted simulating traditional methods at ambient temperatures using lime. The highest indigotin contents were achieved with Indigofera arrecta (55%, Austria) and Strobilanthes cusia (56%, China). There were no statistically significant differences between the indigo extraction yields of the species cultivated in Austria, but Indigofera arrecta and Persicaria tinctoria had statistically significantly higher indigotin extraction yields than Indigofera tinctoria and Indigofera suffruticosa. From the species extracted in China, Strobilanthes cusia showed higher values in all parameters than Indigofera tinctoria, Indigofera suffruticosa and Wrightia laevis. Compared with the standardised method, the method simulating local practice yielded more indigo but had a lower indigotin content; the indigotin extraction yields did not differ greatly. L*a*b* values enabled precise estimations of the indigotin content, making it an interesting option for quality control, as inexpensive, easy-to-handle L*a*b* measuring instruments have become available.


Introduction 1.New Interest in an Ancient Dye
All over the world and for at least 6000 years, natural indigo has been used to dye textiles [1][2][3][4][5].This plant extract, which contains the blue pigment indigotin as the main valuable component for dyeing textiles, became an important globally traded commodity from the 16th century onwards.Natural indigo production reached its peak with colonial mass production on plantations ("indigo factories") between the 17th and 19th centuries, and, like cotton, it was closely linked to the human misery of slavery and other exploitative working conditions.In the late 19th and early 20th centuries, natural indigo was almost completely replaced by synthetic indigo [6,7].What made synthetic indigo initially preferable over the natural product was its constant, standardised quality (indigotin content).Also, later, with the establishment of synthetic indigo production on an industrial scale, its price became cheaper than that of the natural product [8].
Plants 2024, 13, 918 2 of 32 However, natural indigo in the traditional context is still being used today: it has been documented for specific ethnic groups, for example in China [9][10][11][12], India [13][14][15][16][17][18][19] and Indonesia [20,21], but is at risk of disappearing.Nevertheless, there is a renaissance of natural dyes as a sustainable alternative to synthetic dyes and as a means of generating local income in rural areas [22][23][24].As the only plant-derived dye material to achieve longlasting blues (and, in combination with other dyes, greens, violets and blacks as well [25]), natural indigo is regaining its significance.Apart from its use as a textile colourant, natural indigo ("indigo naturalis" in the Chinese Pharmacopoeia) is still used in traditional Chinese medicine for its anti-tumour, anti-inflammatory, anti-psoriatic, antibacterial and antiviral effects [26][27][28][29].
In addition to the blue pigment indigotin, natural indigo can contain other indigoid components (such as the isomers indirubin, cis-indigotin, isoindigotin and isoindirubin, or the intermediate product isatin), flavonoids, and plant-derived and mineral impurities [25,30,31].The impurities result from particulate and colloidal materials from the leaf, dust and soil particles on the leaf surface, or from the use of lime during extraction [32,33].While usually one component (indigotin) is considered the valuable component for dyeing, numerous valuable/bioactive components have been identified for medicinal purposes (indigotin, indirubin, tryptanthrin, isorhamnetin, etc. [26,28,29]).

A Cornucopia of Species and Manufacturing Methods
Thirty-one indigo-yielding species, comprising annual and perennial herbs, shrubs, lianas and trees from eight plant families, have been documented [25], but it is likely that there are many more, as the example of several indigo-yielding Isatis species [34,35] shows.
The following species have been very important economically and were introduced to many areas outside their natural distribution [25].
Besides these widespread species, many other indigo sources have been of local significance, such as the tree Wrightia laevis Hook.f.(Apocynaceae, traditionally used in Hainan, China [12]) and the liana Marsdenia tinctoria R.Br.(Asclepiadeae, distributed in South, Southeast and East Asia [13,25]).
Plants do not contain indigotin, but instead indole glycosides as precursors (indican, isatan A, isatan B and others; some not yet identified) [25].During the manufacturing process, the precursors are transformed into indoxyl, leading to the formation of the blue pigment indigotin and, in a side reaction, the reddish component indirubin (Figure 1).Natural indigo is mostly made by soaking the fresh plant material in water, removing the plants after a certain period, and then aerating the soaking liquid.After the precipitation of the suspended particles, the supernatant is siphoned off.The precipitate can be kept in a liquid state ("indigo paste") or processed further (e.g., boiled, filtered off, pressed) and dried.Dry indigo is either sold solid ("indigo blocks", "indigo cakes") or as finely ground powder [25,33].
Plants 2024, 13, 918 3 of 32 Indigo manufacturing processes differ mainly with regard to their extraction temperature and duration.While most historically documented methods [7,25,51] and traditional methods still practised in Asia extract at ambient temperatures, there are also methods involving higher temperatures that enable quicker extractions (detailed references in Section 3).Another crucial difference concerns the use of lime: slaked lime is used in large quantities in traditional manufacturing processes, for example in China [10][11][12] and Northeast India [17,18].Only small quantities (lime water) were used in indigo production in British India at the end of the 19th century, and various other alkali were tested at that time as well [51].The addition of lime increases the pH, accelerates the conversion of indoxyl to indigotin, and functions as a precipitant for the suspended particles after the aerating step [33,51].However, lime or other alkaline substances are not necessary to produce indigo (and indeed have not been used at all by many producers).If not removed via acidic post-treatment, lime reduces the indigotin content through dilution [51].
Plants 2024, 13, x FOR PEER REVIEW 3 of 31 of the suspended particles, the supernatant is siphoned off.The precipitate can be kept in a liquid state ("indigo paste") or processed further (e.g., boiled, filtered off, pressed) and dried.Dry indigo is either sold solid ("indigo blocks", "indigo cakes") or as finely ground powder [25,33].Indigo manufacturing processes differ mainly with regard to their extraction temperature and duration.While most historically documented methods [7,25,51] and traditional methods still practised in Asia extract at ambient temperatures, there are also methods involving higher temperatures that enable quicker extractions (detailed references in Section 3).Another crucial difference concerns the use of lime: slaked lime is used in large quantities in traditional manufacturing processes, for example in China [10][11][12] and Northeast India [17,18].Only small quantities (lime water) were used in indigo production in British India at the end of the 19th century, and various other alkali were tested at that time as well [51].The addition of lime increases the pH, accelerates the conversion of indoxyl to indigotin, and functions as a precipitant for the suspended particles after the aerating step [33,51].However, lime or other alkaline substances are not necessary to produce indigo (and indeed have not been used at all by many producers).If not removed via acidic post-treatment, lime reduces the indigotin content through dilution [51].
With the development of analytical methods, indigo quality (for the use of dyeing) was usually understood as the content of indigotin [51,52].Analytical methods are manifold and there is still no standard, generally accepted method available (cf. critical review by P. John [33]).Furthermore, data in the literature also relate to different points of reference in indigo processing.Some studies have analysed the content of indigotin precursors in fresh leaves (some of these also calculated a probably too optimistic theoretical indigotin yield [54]), and other studies have analysed the content of indigotin, but by sampling at different processing stages (indigo suspension after aeration, centrifuged indigo, indigo paste, indigo powder, reduced indigo or indigo-dyed textile material), and referring either to the leaf material extracted or to the indigo extracted (details and references are provided in Section 3).In practice, extraction losses occur due to (a) the retention of indoxyl in the leaf tissue during the soaking of the plant material, (b) un-precipitated indigo that is lost when the supernatant is decanted, and (c) the conversion of indoxyl into components other than indigotin, such as indirubin (Figure 1) [33,51].
With the development of analytical methods, indigo quality (for the use of dyeing) was usually understood as the content of indigotin [51,52].Analytical methods are manifold and there is still no standard, generally accepted method available (cf. critical review by P. John [33]).Furthermore, data in the literature also relate to different points of reference in indigo processing.Some studies have analysed the content of indigotin precursors in fresh leaves (some of these also calculated a probably too optimistic theoretical indigotin yield [54]), and other studies have analysed the content of indigotin, but by sampling at different processing stages (indigo suspension after aeration, centrifuged indigo, indigo paste, indigo powder, reduced indigo or indigo-dyed textile material), and referring either to the leaf material extracted or to the indigo extracted (details and references are provided in Section 3).In practice, extraction losses occur due to (a) the retention of indoxyl in the leaf tissue during the soaking of the plant material, (b) un-precipitated indigo that is lost when the supernatant is decanted, and (c) the conversion of indoxyl into components other than indigotin, such as indirubin (Figure 1) [33,51].
Therefore, it is difficult to compare species based on the available data, and only a few studies have directly compared two or more species in experimental research [34,35,[55][56][57][58][59].

Aims of the Study
The objective of this study was to ascertain how indigo-yielding species differ in terms of indigo quality (indigotin content and colour) and extraction yields.Four well-known species-Indigofera tinctoria, I. suffruticosa, Persicaria tinctoria and Strobilanthes cusia-and two less well-known species-Wrightia laevis and Indigofera arrecta-were selected for this study.To the authors' knowledge, there are no experimental data on the indigo yield and quality of Wrightia laevis.Ethnobotanic research has documented that traditional dyers consider its indigo to be of inferior quality to that of the other species used (own fieldwork in Hainan/China in 2019 and [12]).In contrast, Indigofera arrecta was a very promising indigo source during colonial times.In the early 20th century, when natural indigo producers tried to fight the competition presented by synthetic indigo by enhancing their production, it became a topic of agricultural research in India as a high-yielding species [8].However, there is a lack of recent scientific data for this species as well.
A standardised mobile mini extraction (MME) method was developed and used for extractions during field research in China and subsequently for extractions from species grown in the research facility in Austria when field research was no longer possible as a result of the COVID-19 pandemic.A total of 69 samples were provided for spectrophotometric analysis.Wherever possible, indigo was analysed from two to three seed provenances per species.To consider extraction losses and achieve results closer to those in practice, the "final product" (indigo powder) was chosen instead of products at intermediate indigo-processing stages.The results are presented individually for each country since environmental differences between Austria and China did not allow a direct comparison of all species.
Additionally, 25 indigo samples from local producers in China, India and Indonesia were analysed to document and quantify the qualities achieved in practice.To demonstrate the effect of different extraction practices, two species (Persicaria tinctoria and Indigofera arrecta) were used to compare the standardised "hot"-extraction method without lime (MME) and a simulated local practice method at ambient temperatures using lime (sLPE).
Furthermore, the colours of the indigo samples were visually classified, measured (L*a*b* values) and correlated with indigotin contents to establish whether colour could still function as a quality indicator.

Definitions of Terms
The term "indigo" is applied in a confusing way and has different meanings not only in everyday language, but also in the scientific literature as well [60]."Indigo" denominates the following: In the scientific literature dealing with both the plant extract and valuable component analysed, a variety of terms is used to differentiate the extract from the component, but often, the term "indigo" is used for both, and a precise meaning can be established only by understanding the context.Several publications explicitly mention the term "indigotin" as a synonym for "indigo" (Table A1).To add to the confusion, "indigotin" (also "indigotine") can have a different meaning in the context of food chemistry, where it is a synonym for the blue food colourant E 132 (indigo carmine, C 16 H 8 N 2 Na 2 O 2 S 2 ) [61,62].
In this study, the terminology of natural dye handbooks [25,51,63] is followed, and the term "indigo" is used exclusively for the plant extract and "indigotin" for its valuable component for dyeing.Where it is necessary to differentiate between the plant-derived and synthetic products, "natural indigo" and "synthetic indigo", respectively, are used.

Species Cultivated in Austria
The four species cultivated in Vienna, Austria differed significantly from each other in their indigotin content.The highest content was achieved with Indigofera arrecta indigo (55% indigotin).Its amount was double and five times as much, respectively, as those of the two other Indigofera species (Indigofera tinctoria: 23%, Indigofera suffruticosa: 10% indigotin).The indigotin content of Persicaria tinctoria indigo was 44% (Figure 2, upper case letters).
The indigo extraction yields of the species did not differ significantly.The mean values of the provenances per species ranged from 1.3 g/kg (Indigofera suffruticosa) to 1.6 g/kg (Indigofera tinctoria).
The extraction yields of indigotin, however, differed significantly between two groups of species: Persicaria tinctoria and Indigofera arrecta (0.6 and 0.8 g/kg) showed much higher indigotin extraction yields than Indigofera suffruticosa and Indigofera tinctoria (0.1 and 0.3 g/kg).As there was no statistically significant difference in the indigo extraction yields, these two groupings must mainly be caused by the indigotin content.
Comparing different seed provenances within a species (Figure 2, lower case letters), the Persicaria tinctoria provenances showed statistically significant differences in almost all the parameters tested.In contrast, the Indigofera arrecta provenances did not differ significantly, but the data varied widely.The same was true for the provenances of Indigofera tinctoria and-apart from differences in indigotin content-for Indigofera suffruticosa as well.The results for I. tinctoria and I. suffruticosa, however, may be additionally influenced by the slightly different composition of the plant material extracted from one provenance of each species (Section 4).

Species Cultivated in China
The results from the extractions during fieldwork in three villages in China were evaluated separately from those cultivated and extracted in Austria owing to differences in species, environmental conditions and cultivation practices (Figure 3).In China, apart from in the case of Wrightia laevis (harvested from two trees, both growing in the same village), it was possible to extract only one provenance per species.
Compared with the other species extracted in China, Strobilanthes cusia showed the highest values for all the parameters tested.The differences between Strobilanthes cusia and the other species were also statistically significant in most cases (apart from the indigo extraction yield compared with Indigofera suffruticosa and the indigotin content compared with tree 1 of Wrightia laevis).Due to the high indigo extraction yield combined with a high indigotin content, the indigotin extraction yield of Strobilanthes cusia was up to five times greater than that of the other species (Figure 4).
highest values for all the parameters tested.The differences between Strobilanthes cusia and the other species were also statistically significant in most cases (apart from the indigo extraction yield compared with Indigofera suffruticosa and the indigotin content compared with tree 1 of Wrightia laevis).Due to the high indigo extraction yield combined with a high indigotin content, the indigotin extraction yield of Strobilanthes cusia was up to five times greater than that of the other species (Figure 4).highest values for all the parameters tested.The differences between Strobilanthes cusia and the other species were also statistically significant in most cases (apart from the indigo extraction yield compared with Indigofera suffruticosa and the indigotin content compared with tree 1 of Wrightia laevis).Due to the high indigo extraction yield combined with a high indigotin content, the indigotin extraction yield of Strobilanthes cusia was up to five times greater than that of the other species (Figure 4).

Influence of Extraction Methods (MME, sLPE and Modifications) on Indigo Quality and Yield
The standardised method of extracting the plant material for three hours at 45 °C without using lime (MME) was compared with another extraction method (sLPE) that simulates local extraction practices at ambient temperatures with the addition of lime before the aeration step.The influence of these two methods was demonstrated using two provenances of

Influence of Extraction Methods (MME, sLPE and Modifications) on Indigo Quality and Yield
The standardised method of extracting the plant material for three hours at 45 °C without using lime (MME) was compared with another extraction method (sLPE) that simulates local extraction practices at ambient temperatures with the addition of lime before the aeration step.The influence of these two methods was demonstrated using two provenances of Values with the same letter do not differ significantly (ANOVA; p > 0.05).

Influence of Extraction Methods (MME, sLPE and Modifications) on Indigo Quality and Yield
The standardised method of extracting the plant material for three hours at 45 • C without using lime (MME) was compared with another extraction method (sLPE) that simulates local extraction practices at ambient temperatures with the addition of lime before the aeration step.The influence of these two methods was demonstrated using two provenances of Indigofera arrecta and three of Persicaria tinctoria, all cultivated in Austria.The extraction using lime (sLPE) yielded more indigo, but it had a much lower indigotin Plants 2024, 13, 918 9 of 32 content than the indigo from the extraction without lime (MME).This was unsurprising since without any post-treatment the lime added during extraction remains in the extract and therefore dilutes the product.The differences in all the parameters analysed were statistically significant, except for the indigotin extraction yield of Indigofera arrecta (Table 1).
Further modifications of the extraction methods (mod.MME, mod.sLPE) were tested with one provenance of Indigofera arrecta to see whether there was scope for improvement.The extraction durations of both modified methods were determined using the "state" of the soaking liquid, recognisable by its change in appearance and smell, as was performed during colonial indigo production and is still practised today.Concerning the modified MME method, the soaking duration extended to five hours (rather than three hours), successfully increased both the indigo and indigotin extraction yields compared to the original MME method, even though the content of indigotin in the indigo remained the same (58%).The original sLPE method yielded rather greenish indigo with Indigofera arrecta, which was interpreted as low quality.Furthermore, the smell of the liquid at the end of the soaking period was already putrid, indicating "excessive fermentation".The reduction of the soaking duration from 48 h to 28 h, which was determined by the appearance and smell of the liquid, did yield blue indigo with mod.sLPE, but the indigotin extraction yield remained more or less the same as with the original sLPE method.The increase in indigo yield combined with a decrease in indigotin content of the mod.sLPE method can be explained by the use of more lime (Table 1).project partners in India and Indonesia.Due to the different species, sites and extraction methods, the sample set was too heterogeneous and too small to be analysed statistically for differences in indigotin content, but it was very useful for providing comprehensive, quantitative documentation of indigo qualities achieved in practice and for a first test of the L*a*b* method.
Locally produced indigo from Indigofera tinctoria had an indigotin content of 2-23% (Figure 5a).The maximum content was achieved through extractions without using lime in India and matched the indigotin contents achieved with the standardised extractions of this species in Austria (23%; Figure 2) and China (23%; Figure 4).
The indigotin content of locally produced indigo from Strobilanthes cusia ranged from 4 to 36% (Figure 5b).Again, the maximum indigotin content was from local extractions without the use of lime in India.However, the maximum remained 20% below that achieved using standardised extraction (56%; Figure 4).
The only sample of Wrightia laevis indigo obtained in accordance with local practice had an indigotin content of 2% (extraction with lime, Xin Cun, China).
Plants 2024, 13, x FOR PEER REVIEW 10 of 31 In total, 25 indigo samples from local extraction practices were purchased from traditional small-scale producers in China during field research with local partners, and from project partners in India and Indonesia.Due to the different species, sites and extraction methods, the sample set was too heterogeneous and too small to be analysed statistically for differences in indigotin content, but it was very useful for providing comprehensive, quantitative documentation of indigo qualities achieved in practice and for a first test of the L*a*b* method.
Locally produced indigo from Indigofera tinctoria had an indigotin content of 2-23% (Figure 5a).The maximum content was achieved through extractions without using lime in India and matched the indigotin contents achieved with the standardised extractions of this species in Austria (23%; Figure 2) and China (23%; Figure 4).
The indigotin content of locally produced indigo from Strobilanthes cusia ranged from 4 to 36% (Figure 5b).Again, the maximum indigotin content was from local extractions without the use of lime in India.However, the maximum remained 20% below that achieved using standardised extraction (56%; Figure 4).
The only sample of Wrightia laevis indigo obtained in accordance with local practice had an indigotin content of 2% (extraction with lime, Xin Cun, China).The visual classification of the indigo samples showed a colour range from dark blue with a violet tinge to greenish shades when extractions with the sLPE method did not work properly.Light and mid blue shades occurred when lime was used during the extraction.In total, eight colour classes were identified, with most of the samples in the class "dark blue/violet" (Table 2; data of all samples: Tables S1 and S2).extraction methods were roughly classified into methods using lime and methods without lime.For the number of samples per village indicated (n), "n = 1 + 1" means two samples from the same extraction, one sample from the first processing without lime and one sample from the further processing using lime.

Visual Colour Classification of the Indigo Samples
The visual classification of the indigo samples showed a colour range from dark blue with a violet tinge to greenish shades when extractions with the sLPE method did not work properly.Light and mid blue shades occurred when lime was used during the extraction.In total, eight colour classes were identified, with most of the samples in the class "dark blue/violet" (Table 2; data of all samples: Tables S1 and S2).The species differed significantly with regard to the L*a*b* values of the indigo (MME samples, Table A4).Within the provenances of each species, however, in most cases, there were no statistically significant differences (Table A5).
The positions of the indigo samples colours in the colour space (Figure 6) can easily be understood by imagining their colours described with the visual colour classification.For example, samples visually classified as "dark blue/violet" (such as all the samples extracted using the MME method from Indigofera arrecta, Persicaria tinctoria and Strobilanthes cusia, Table S1) are positioned between blue (negative b*-axis) and red (positive a*-axis), further away from the centre.As they are dark, they have low L* values.Blues without a reddish tinge ("dark blue", "mid blue", "light blue") are positioned further away from red (negative a*): the light and mid blue samples (such as the sLPE samples from Persicaria tinctoria or the mod.sLPE samples from Indigofera arrecta) are further away from the centre and have higher L* values than the dark blue samples (such as the MME samples of Indigofera suffruticosa positioned further left)."Green" samples (from malfunctioning sLPE extractions of Indigofera arrecta) are positioned in the area between green and yellow (negative a* and positive b* axes), almost opposite the dark blue/violet colour (Figure 6a).

Measured Colour Parameters (L*a*b*)
The species differed significantly with regard to the L*a*b* values of the indigo (MME samples, Table A4).Within the provenances of each species, however, in most cases, there were no statistically significant differences (Table A5).
The positions of the indigo samples' colours in the colour space (Figure 6) can easily be understood by imagining their colours described with the visual colour classification.For example, samples visually classified as "dark blue/violet" (such as all the samples extracted using the MME method from Indigofera arrecta, Persicaria tinctoria and Strobilanthes cusia, Table S1) are positioned between blue (negative b*-axis) and red (positive a*-axis), further away from the centre.As they are dark, they have low L* values.Blues without a reddish tinge ("dark blue", "mid blue", "light blue") are positioned further away from red (negative a*): the light and mid blue samples (such as the sLPE samples from Persicaria tinctoria or the mod.sLPE samples from Indigofera arrecta) are further away from the centre and have higher L* values than the dark blue samples (such as the MME samples of Indigofera suffruticosa positioned further left)."Green" samples (from malfunctioning sLPE extractions of Indigofera arrecta) are positioned in the area between green and yellow (negative a* and positive b* axes), almost opposite the dark blue/violet colour (Figure 6a).
Considering the results from the indigotin analysis (Figures 2a and 4a), it seems that the species that showed high indigotin contents in their indigo when extracted using the MME method (Indigofera arrecta, Persicaria tinctoria and Strobilanthes cusia) are positioned closer to the synthetic indigo standard, whose colour is also a dark blue with a violet tinge.Samples with a lower indigotin content, especially those from sLPE extractions, are positioned further away from the synthetic indigo standard and are lighter (high L* values) due to the remaining lime in the indigo.
Only a few samples from local practice extractions (Figure 6b) show positive a* and negative b* values ("dark blue/violet").They are not as close to the synthetic standard as some samples from the MME extractions.Most of the local practice samples are positioned in the area between the negative b* and negative a* axes and show "dark blue/black", "dark blue" and "mid blue" colours.The sample with by far the highest L* value and that is furthest from the centre is "light blue" (Wrightia laevis).
the species that showed high indigotin contents in their indigo when extracted using the MME method (Indigofera arrecta, Persicaria tinctoria and Strobilanthes cusia) are positioned closer to the synthetic indigo standard, whose colour is also a dark blue with a violet tinge.Samples with a lower indigotin content, especially those from sLPE extractions, are positioned further away from the synthetic indigo standard and are lighter (high L* values) due to the remaining lime in the indigo.
Only a few samples from local practice extractions (Figure 6b) show positive a* and negative b* values ("dark blue/violet").They are not as close to the synthetic standard as some samples from the MME extractions.Most of the local practice samples are positioned in the area between the negative b* and negative a* axes and show "dark blue/black", "dark blue" and "mid blue" colours.The sample with by far the highest L* value and that is furthest from the centre is "light blue" (Wrightia laevis).The correlation between the measured colour coordinates and the indigotin content was analysed using data from the standardised extractions performed in Austria and China (MME and sLPE).For both extraction methods, there was a strong negative correlation between indigotin content and the L* value.Furthermore, the two extraction methods differed greatly in terms of their correlation patterns: while MME samples showed a strong positive correlation between indigotin content and a* values, in sLPE samples, the indigotin content and a* values were not correlated.MME samples showed a strong positive correlation between a* and L* values, whereas sLPE samples showed a strong negative correlation between a* and b* values (Figure 7).
two axes of the complementary colours green-red (a* axis) and blue-yellow (b* axis); the third dimension, lightness (L* axis, for black-white), is indicated by the sizes of the data points (continuous scale); bigger sizes indicate higher L* values (i.e., lighter colours), smaller sizes indicate lower L* values (darker colours).(a) Samples from standardised extractions (three replications); the small, inserted graph shows the maximum range of the axes according to the Free Colour Converter provided by Nix Sensor Ltd., Hamilton, Canada (www.nixsensor.com,accessed 27 June 2023); (b) samples from local practice extractions.

Correlation between Indigo Colour (L*a*b* Values) and Indigotin Content
The correlation between the measured colour coordinates and the indigotin content was analysed using data from the standardised extractions performed in Austria and China (MME and sLPE).For both extraction methods, there was a strong negative correlation between indigotin content and the L* value.Furthermore, the two extraction methods differed greatly in terms of their correlation patterns: while MME samples showed a strong positive correlation between indigotin content and a* values, in sLPE samples, the indigotin content and a* values were not correlated.MME samples showed a strong positive correlation between a* and L* values, whereas sLPE samples showed a strong negative correlation between a* and b* values (Figure 7).3).L*, a* and the interaction between L* and a* predicted the indigotin content significantly (all p < 0.01).3).L*, a* and the interaction between L* and a* predicted the indigotin content significantly (all p < 0.01).

Indigotin Content and Indigo(tin) Extraction Yield
The lack of a standardised method for determining indigotin and the wide variety of reference points used in previous studies (dry indigo, indigo paste, indigo suspension, etc.) allow only a limited comparison of results.Comparable data are often not available, even for a widely used species such as Indigofera tinctoria.An analysis of extraction yields with those of previous studies is even more limited because data also refer to different compositions of the plant material extracted (leaves and stems versus leaves only) and sometimes refer to the dry weight of plant material, even though the extraction was Plants 2024, 13, 918 14 of 32 performed using fresh plant material (Table 4).The composition of the plant material extracted is of relevance because in most species, indigo precursors are present only in the leaves [25].However, it has been demonstrated with Strobilanthes cusia that even stems contain precursors that can be extracted in a separate process using specific methods [49].
The highest indigotin content maxima in indigo documented in previous studies (including additional species not analysed in the present study; Table 4) are about 40%.The highest indigotin content (57%) achieved was in just one study [50], in which Strobilanthes cusia was extracted in an elaborated two-step method using nitrogen gas instead of oxidation by air, and the indigo was subsequently purified by removing indirubin with acetone extraction.In most studies, the indigotin content maxima barely reached 15%.The results of the present study, however, were in most cases far above 15%, and the highest values (58% with two Indigofera arrecta provenances and 56% with Strobilanthes cusia) matched the maxima of earlier studies, even though the MME method is a much simpler method than that used by Jin et al. [50].However, there might still be plenty of scope for improvement in view of the indigotin contents documented in 19th-century sources: indigos with less than 20% indigotin were classified as the lowest qualities, while the best indigo qualities contained 70-90% indigotin [52].Unfortunately, these data do not provide information about the analytical methods or indigo-yielding species used.
With regard to the indigo extraction yield, only one study on Persicaria tinctoria provides results in a form that is more comparable: the indigo extraction yields achieved by Bechtold et al. ranged from 1.0 to 7.9 g/kg, and the maximum reached by one extraction was 13.0 g/kg [54].These results are generally much higher than the maximum achieved in the present study's extractions of Persicaria tinctoria using the MME method (1.7 g/kg, provenance "Kojyoko", Figure 2) but also above the maximum indigo extraction yield of all the species evaluated (4.7 g/kg achieved with Strobilanthes cusia, Figure 4).
The indigotin extraction yields were within the range achieved in most previous studies with comparable reference points, apart from two studies that document much higher indigotin extraction yields achieved with Strobilanthes cusia (4.7 g/kg [59] and 4.5 g/kg [50]) and Indigofera tinctoria (3.3 g/kg [59]).
The seed provenances of the three Indigofera species cultivated and extracted in Austria originated from wild collections, apart from one Indigofera arrecta provenance, for which its origin is unknown (Table A2).As Indigofera tinctoria, I. suffruticosa and I. arrecta have been intensively cultivated for centuries, it is possible that other seed provenances of plants in cultivation produce higher yields due to breeding and/or adaptation to specific sites.Statistically significant differences between different seed provenances concerning precursor or indigotin content and extraction yields have been documented in previous studies with Indigofera tinctoria [58,64,65], Isatis tinctoria [66,67] and Persicaria tinctoria [68], while another study with Persicaria tinctoria could not confirm statistically significant differences [69].To date, studies evaluating Strobilanthes cusia provenances are still lacking (or are not accessible).
In the search for seeds for the cultivation and extraction experiments, several were incorrectly declared to be "Indigofera tinctoria", even in specialised seed trade and botanic gardens.As the genus Indigofera is huge (750 species [70]) and precise species identification is challenging, cooperation with experienced botanists is highly recommended.

The Evaluation of Indigo-Yielding Species Requires a Differentiated Approach
Of the species extracted using the MME method in China, Strobilanthes cusia showed the best results concerning indigotin content and extraction yields.Of the species cultivated and extracted in Austria, Indigofera arrecta had the highest indigotin content and, together with Persicaria tinctoria, the highest indigotin extraction yield (regarding indigo extraction yields, there were no statistically significant differences between the species tested in Austria).
The superiority of Strobilanthes cusia compared with other indigo-yielding species has been highlighted by many traditional producers/dyers interviewed in ethnobotanical studies in China [10,12] and has also been documented in experiments comparing Strobilanthes cusia with Indigofera tinctoria [59].The results of the present study indicate that the superiority of Strobilanthes cusia could be due to a high content of indigotin in the indigo, combined with a high indigo extraction yield.A high indigotin content alone, however, can also be achieved with other species, especially if extraction methods are optimised.Nevertheless, the indigo extraction yield of those species is much lower or, as in the case of Indigofera suffruticosa extracted in China (MME), is also high but the content of indigotin is low.
Therefore, when comparing indigo-yielding species, it is important to differentiate between the perspective of the indigo user (content of indigotin) and the perspective of the indigo producer (indigo and indigotin extraction yield per kg of plant material and also per hectare per year).For medicinal purposes, other components found in indigo are also relevant, and for dyers this may also be the case if minor components contribute to specific shades of blue.
3.3.Indigofera arrecta-A "Forgotten Species" with a Promising Future?Indigo from the three Indigofera arrecta provenances cultivated in Austria showed the same indigotin content as that from Strobilanthes cusia cultivated in Hainan (MME method).The indigotin content and indigotin extraction yield of I. arrecta were much higher than those of the two widely used Indigofera species, I. tinctoria and I. suffruticosa, irrespective of whether these were cultivated in Austria or tropical Hainan.
These promising results explain why Indigofera arrecta, native to tropical and southern Africa and the Arabian Peninsula [71], was widely transferred to indigo-producing areas in the 19th century: first to Java (named "Natal indigo" according to its origin) and then from there-becoming "Java indigo"-to India [2].Due to its high yields, it played an important role in the attempts of natural indigo producers in colonial India to fight the competition presented by synthetic indigo in the late 19th/early 20th century [8].As the present results show, Indigofera arrecta could again become an interesting alternative source of indigo in (sub)tropical regions.Its potential use, however, might not be restricted to these areas only.
In previous studies searching for indigo sources suitable for production in temperate regions, Persicaria tinctoria showed better results than Isatis tinctoria with regard to precursor or indigotin content and extraction yields, both at the laboratory scale [56] and pilot/prototype scale ( [55], Table 4).In the present experiments (MME), Indigofera arrecta indigo had a 10% higher indigotin content than Persicaria tinctoria, and its indigo and indigotin extraction yields did not differ significantly from those of Persicaria tinctoria.
Being a plant whose natural habitats are moist valleys and stream banks [72], Persicaria tinctoria has been cultivated in humid regions, such as in Tokushima Prefecture, Japan [25,42] (according to the updated Köppen-Geiger climate classification, Tokushima is in a warm temperate climate zone, fully humid (i.e., rain at all seasons) and with hot summers [73]).Otherwise, Persicaria tinctoria requires irrigation as it is sensitive to drought stress [56].With ongoing climate change, the irrigation of field crops will become increasingly challenging.Indigofera arrecta-which, as demonstrated here, can be grown as an annual crop in temperate climate conditions-is possibly more tolerant to dry conditions than Persicaria tinctoria.However, this still needs to be proven in experimental research, as data concerning its requirements of minimum annual precipitation [74] are rare and contradictory, and there is an absence of data concerning the annual distribution of precipitation.Another potential advantage of Indigofera arrecta compared with Persicaria tinctoria concerns nutrient supply: Persicaria tinctoria requires intense fertilising (cultivation instructions for conventional farming [75] recommend 150-200 kg N/ha), while Indigofera arrecta as a leguminous species has the advantage of symbiotic nitrogen-fixing bacteria and positive effects in crop rotation [74].

The Ambivalence of Using Standardised Methods, and Potential Improvements
As a first step, the standardised methods were useful for comparing indigo-yielding species, but they also represent a simplification and compromise in terms of speciesspecific requirements.A standardised extraction method does not mean an optimal extraction method.
The MME method was developed as a transportable field method in pre-experiments with Persicaria tinctoria and adapted after field testing with the same species and Strobilanthes cusia.For all species except Wrightia laevis, this method worked well.With Wrightia laevis, the standardised extraction duration would have yielded no indigo at all, but when the soaking duration was extended, neither the indigotin content nor extraction yield were significantly different from that of Indigofera tinctoria or I. suffruticosa.
In most species, precursors are present in the leaves.The leaves of indigo-yielding species vary in size and consistency, so it could be that the "extraction behaviour" is species specific.Furthermore, other factors may be also species specific, such as the types of precursors, the microbiome and enzymes of the plants [25], or there are additional substances present in leaves that could have an impact on the process (Wrightia laevis, for example, is the only species tested that contains a milky sap, as observed during extractions).
In colonial times, indigo producers noticed that it was necessary to adapt the soaking time, in relation not only to different species, but also to the age and storage duration of the harvested plants before extraction and to the ambient temperature during extraction [52].Practitioners have always been aware of this and have defined durations of processing steps rather by the observation of visible, olfactory or haptic changes during the process (such as a change in the colour or smell of the extraction liquid, the occurrence of bubbles, etc.) [7,10,12,52,76].

Colour Measurement as a Possible New Method to Determine Indigo Quality
The results of the present study confirm the relationship between indigo colour and indigo quality, as is already known from practice [51][52][53].The regression analysis of the L*a*b* colour coordinates and the indigotin content of indigo samples determined spectrophotometrically showed that it is possible to predict the content of indigotin, provided that all three colour coordinates, including their possible interactions, are considered in the calculation.Improved standardisation with more exact laboratory methods (e.g., HPLC) could result in a better prediction of indigotin.Since cheap handheld L*a*b* measurement devices are available and their measurement of indigo is fast without tedious sample preparation, it may become an interesting option for routine analyses.

Materials and Methods
Standardised extractions yielding 69 indigo samples were performed during ethnobotanical field research in China with locally cultivated species and with species cultivated in Austria (Indigofera tinctoria, I. suffruticosa, I. arrecta, Persicaria tinctoria, Strobilanthes cusia and Wrightia laevis).In addition, 25 local indigo samples from indigo producers in China, India and Indonesia were analysed (Indigofera tinctoria, Strobilanthes cusia and Wrightia laevis).In total, 94 indigo samples from six species were analysed (Tables 5, S1 and S2).The number of indigo samples per species varied between 7 and 23.MME procedure: Extractions were performed in three replications simultaneously with plant material from the same harvest (mixed sample from various individual plants).Immediately after the harvesting, 500 g of fresh plant material was rinsed with water and distributed equally into two mesh bags.The plant material was soaked for 3 h in three litres of water at 45 • C (the plants were kept submersed by two tightly closed glasses filled with water and placed on top of the mesh bags).The mesh bags with the plant material were then taken out and squeezed to avoid loss of the soaking liquid.The liquid was aerated for 30 min with a fish tank pump and then left for 20 h to allow the pigment to settle at the bottom.The whole liquid was then repeatedly filtered off through white cotton fabric until the filtrate was clear.
Modified MME of Indigofera arrecta: Thick stems were not removed, and the plant material was soaked until a thick copperish skin became visible on the surface, but the smell was still a "good indigo smell", as described by P.-P.Darrac [7], before the typical putrid smell of excessive fermentation was noticeable (MME.69-71,unknown provenance: soaking duration, 5 h).
Modified MME of Wrightia laevis: As the standardised soaking duration of 3 h would not have yielded any indigo, the soaking duration was extended until a copperish film became visible on the surface of the liquid (MME.13-15,tree 1: soaking duration, 6 h; MME.16-18, tree 2: 9 h).

sLPE (Simulated Local Practice Extraction)
This extraction method was carried out only with Persicaria tinctoria and Indigofera arrecta.It was developed to simulate traditional practices of extraction at ambient temperature and the addition of lime before aeration.The method builds on previous experiments of the authors with Persicaria tinctoria [84] and Strobilanthes cusia [85].sLPE procedure: Extractions were performed in three replications and carried out simultaneously with plant material from the same harvest as used for the MMEs.Immediately after the harvesting, 500 g of fresh plant material was rinsed with water and distributed equally into two mesh bags.The mesh bags with the plant material were placed around a perforated pipe in the centre of a beaker.Three litres of water at 20 • C were added, and the beaker was kept in an incubator at 30 • C for 48 h (the plants were kept submersed by two tightly closed glasses filled with water and placed on top of the mesh bags).The mesh bags with the plant material were then taken out and squeezed to avoid loss of the soaking liquid.Then, 5 g of hydrated lime (Ca(OH) 2 ) mixed in a small amount of water was added to the soaking liquid (a constant amount of lime rather than a standardised pH value was used in order to avoid influencing the indigo yield with the weight of lime added).The liquid was aerated for 30 min with a fish tank pump and afterwards left for 20 h to allow the pigment to settle at the bottom.The whole liquid was then repeatedly filtered off through white cotton fabric until the filtrate was clear.
Modified sLPE of Indigofera arrecta: Thick stems were not removed, and the plant material was soaked until a thick copperish skin became visible on the surface and the smell was still a "good indigo smell", before the typical putrid smell of an excessive fermentation was noticeable (soaking duration: 28 h instead of 48 h).Instead of 5 g of lime, 7 g of lime were used to adjust to a pH of 11 (sLPE.28-30,unknown provenance).

Indigo Samples from Local Practices in China, Indonesia and India
Indigo samples were bought from traditional local indigo producers during field research in China in 2019 and from indigo producers in India and Indonesia in 2021, with their permission to analyse and publish results (prior informed consent).
LPE (local practice extraction) samples were extracted by local people during field research in China at the same time and using the same plant material of Indigofera tinctoria, Strobilanthes cusia and Wrightia laevis from the same harvest as the corresponding MMEs.The traditional extraction method uses lime during the aeration of the soaking liquid.In principle, the method is very similar regardless of the species on which the extraction is being performed; its variations mainly concern the duration of the processing steps (Tables 5 and S2).
ADD samples (additional samples from local practice) were bought during field research in China and taken from indigo already available in the villages from previous harvests (species according to the interviewees, traditional extraction method).Samples were also obtained from project partners in India (Avani) and Indonesia (Threads of Life) (extraction methods with and without using lime, Tables 5 and S2).

Indigo Sample Preparation
The cotton filters with the indigo pigment were first air-dried, and then dried in the drying cabinet according to the standardised procedure at 60 • C until weight was consistent.The dry indigo was scraped off the filter and kept protected from light (using brown glass vials in a cardboard box).The samples received from Avani, India (traded as powder) and from Threads of Life, Indonesia (normally traded as paste, but air-dried and roughly ground for shipment to Vienna) were not dried in a drying cabinet to avoid cross-contamination.All the samples were ground with a Qiagen Tissue Lyser 2 (Venlo, The Netherlands), with the samples in 2 mL microcentrifuge safe-lock tubes (Eppendorf, Hamburg, Germany) with four zirconium oxide grinding balls (3 mm) per tube (frequency: 30 Hz, grinding duration: 2:30 min).The spectrophotometric method of Bechtold et al. [54,86] measures the content of leucoindigotin in an aqueous alkaline solution with Fe(II)-triethanolamine as the reducing agent.The method was adapted to a 96-well microplate reader format, and measurements were performed in non-treated 96-well microplates (BRANDplates ® pureGrade™, Wertheim, Germany) using a plate reader (Infinite M200Pro, Tecan, Männedorf, Switzerland).
In a volumetric flask (100 mL), ground indigo (5 mg) was reduced with 20 mL of solution A (33 g/L NaOH, 333 g/L triethanolamine, 50 g/L FeSO 4 •7H 2 O, 0.8 g/L citric acid) and filled up to 100 mL with solution B (10 g/L NaOH, 50 g/L triethanolamine, 5 g/L FeSO 4 •7H 2 O).The volumetric flask with a stopper was kept in a water bath at 40 • C for 60 min (shaking every 15 min, three times in total).For each indigo sample, three sample preparations in duplicate (150 µL per well) were analysed on different days at 40 • C (microplate with lid, waiting for 2 min after insertion to avoid steaming up), and with absorbance at a wavelength of 410 nm.Per analysis run, six standards (0-50 ppm) and a maximum of ten indigo samples were analysed to ensure the completion of the analyses within the given time frame for sample preparation stability (2 h).Synthetic indigo (Acros Organics, ≥94%) was used as the standard for calibration.
Intra-day and inter-day variations for the MME method were 1.4% and 5.42%, respectively, and for the sLPE method, 5.4% and 16.4%, respectively.The regression coefficient was 0.9945.

Conclusions
This study provides an evaluation of six indigo-yielding species (Indigofera arrecta, I. tinctoria, I. suffruticosa, Persicaria tinctoria, Strobilanthes cusia and Wrightia laevis) with regard to indigo quality (content of indigotin) as well as indigo and indigotin extraction yields.A comparison of indigo-yielding species requires a differentiated approach: a high indigotin content in the indigo can be achieved with several species (the maxima were 55% with Indigofera arrecta, Austria, and 56% with Strobilanthes cusia, China).Even with Wrightia laevis, a species known for low quality, a similar indigotin content to that from the Indigofera species can be achieved when the extraction methods are optimised.The indigotin yield per kg of plant material, however, was clearly highest with Strobilanthes cusia (2.6 g/kg) compared with 0.5-0.7 g/kg extracted in China from the other species (without statistically significant difference among Indigofera tinctoria, I. suffruticosa and Wrightia laevis) and the maxima achieved in Austria (0.6 and 0.8 g/kg achieved with Persicaria tinctoria and Indigofera arrecta, respectively).
Indigofera arrecta, a species once highly valued, could become an interesting alternative species again due to its high indigo quality and indigotin extraction yield.
Indigo colour, once (and, in some traditional contexts, still) used as an indicator of quality, could still serve this purpose in a broader context as well, as demonstrated by the first correlation and regression analysis.As small, inexpensive, and easy-to-handle L*a*b* measurement tools are now available, this method could become a quick quality control tool both in scientific field research and in practice as well, without the need for costly laboratory facilities.

Figure 1 .
Figure 1.The blue pigment indigotin and the reddish component indirubin.

Figure 1 .
Figure 1.The blue pigment indigotin and the reddish component indirubin.
(a) The plants used; (b) The extract as a traded product; (c) The valuable component for dyeing contained in the extract; (d) The blue colour.

Figure 2 .
Figure 2. Comparison of species and seed provenances within each species, cultivated and extracted using the MME method in Austria: (a) content of indigotin in the indigo, (b) indigo extraction yield and (c) indigotin extraction yield.Values with the same letter do not differ significantly (ANOVA, p > 0.05; upper case letters: comparison of species; lower case letters: comparison of seed provenances within a species).

Figure 2 .
Figure 2. Comparison of species and seed provenances within each species, cultivated and extracted using the MME method in Austria: (a) content of indigotin in the indigo, (b) indigo extraction yield and (c) indigotin extraction yield.Values with the same letter do not differ significantly (ANOVA, p > 0.05; upper case letters: comparison of species; lower case letters: comparison of seed provenances within a species).

Figure 3 .
Figure 3. Composition of harvested plant material from Indigofera species cultivated in China and Austria.(a,b): Indigofera tinctoria from Yu Long, China.(c-e): Indigofera suffruticosa, provenance from Ecuador, cultivated in Austria.Pictures: A. Hartl.

Figure 3 .
Figure 3. Composition of harvested plant material from Indigofera species cultivated in China and Austria.(a,b): Indigofera tinctoria from Yu Long, China.(c-e): Indigofera suffruticosa, provenance from Ecuador, cultivated in Austria.Pictures: A. Hartl.

Figure 3 .
Figure 3. Composition of harvested plant material from Indigofera species cultivated in China and Austria.(a,b): Indigofera tinctoria from Yu Long, China.(c-e): Indigofera suffruticosa, provenance from Ecuador, cultivated in Austria.Pictures: A. Hartl.

Figure 4 .
Figure 4. Comparison of locally cultivated species extracted using the MME method in China: (a) content of indigotin in the indigo, (b) indigo extraction yield and (c) indigotin extraction yield.Values with the same letter do not differ significantly (ANOVA; p > 0.05).

Figure 4 .
Figure 4. Comparison of locally cultivated species extracted using the MME method in China: (a) content of indigotin in the indigo, (b) indigo extraction yield and (c) indigotin extraction yield.Values with the same letter do not differ significantly (ANOVA; p > 0.05).

Figure 4 .
Figure 4. Comparison of locally cultivated species extracted using the MME method in China: (a) content of indigotin in the indigo, (b) indigo extraction yield and (c) indigotin extraction yield.Values with the same letter do not differ significantly (ANOVA; p > 0.05).

Figure 5 .
Figure 5. Content of indigotin achieved using local extraction practices.Indigo samples: (a) Indigofera tinctoria and (b) Strobilanthes cusia.Location is indicated by country and by village.The

Figure 5 .
Figure 5. Content of indigotin achieved using local extraction practices.Indigo samples: (a) Indigofera tinctoria and (b) Strobilanthes cusia.Location is indicated by country and by village.The extraction methods were roughly classified into methods using lime and methods without lime.For the number of samples per village indicated (n), "n = 1 + 1" means two samples from the same extraction, one sample from the first processing without lime and one sample from the further processing using lime.

2. 4 .
Is Indigo Colour an Indicator of Indigo Quality? 2.4.1.Visual Colour Classification of the Indigo Samples

Table 2 .
Examples representing the eight visually determined colour classes (including sample information and L*a*b* values of the samples shown in the picture) and the number of samples per colour class.

Table 2 .
Examples representing the eight visually determined colour classes (including sample information and L*a*b* values of the samples shown in the picture) and the number of samples per colour class.

Figure 6 . 3 .
Figure 6.Positions of indigo samples in the colour space according to their L*a*b* coordinates, with synthetic indigo as the reference point.The three-dimensional CIELAB colour space is shown by the two axes of the complementary colours green-red (a* axis) and blue-yellow (b* axis); the third dimension, lightness (L* axis, for black-white), is indicated by the sizes of the data points (continuous scale); bigger sizes indicate higher L* values (i.e., lighter colours), smaller sizes indicate lower L* values (darker colours).(a) Samples from standardised extractions (three replications); the small, inserted graph shows the maximum range of the axes according to the Free Colour Converter provided by Nix Sensor Ltd., Hamilton, Canada (www.nixsensor.com,accessed 27 June 2023); (b) samples from local practice extractions.

Figure 7 .
Figure 7. Correlation between the indigotin content and L*a*b* values of indigo samples: (a) extraction with the MME method and (b) extraction with the sLPE method.Pearson correlation coefficients.The aggregated data set was used.2.4.4.Can Indigotin Content Be Predicted by Measuring Indigo Colour?Multiple linear regression was used to test if all L*a*b* values and their interactions together could significantly predict the indigotin content.The overall regression was statistically significant (R 2 = 0.89, F (7, 60) = 66.09, p < 0.001, Table3).L*, a* and the interaction between L* and a* predicted the indigotin content significantly (all p < 0.01).

Figure 7 .
Figure 7. Correlation between the indigotin content and L*a*b* values of indigo samples: (a) extraction with the MME method and (b) extraction with the sLPE method.Pearson correlation coefficients.The aggregated data set was used.2.4.4.Can Indigotin Content Be Predicted by Measuring Indigo Colour?Multiple linear regression was used to test if all L*a*b* values and their interactions together could significantly predict the indigotin content.The overall regression was statistically significant (R 2 = 0.89, F (7, 60) = 66.09, p < 0.001, Table3).L*, a* and the interaction between L* and a* predicted the indigotin content significantly (all p < 0.01).

4. 4
.2. Calculations Content of indigotin: The indigotin content in the indigo pigment (weight percent) was calculated from the single concentration measured and the indigo sample weight.Indigo extraction yield: As the indigo pigment could not be scraped off the filter without losses (indigo remaining on the filter), the indigo yield in g (dry) per kg fresh plant material was calculated by determining the dry weight of the filter with indigo and subtracting the filter weight (both drying at 60 • C, until weight consistency).Indigotin extraction yield: The indigotin yield in g (dry) per kg of fresh plant material was calculated based on the indigo extraction yield and the indigotin content of the sample.

4. 5 .
Determination of Indigo Sample Colour 4.5.1.Visual Classification of ColoursAll standardised ground indigo samples were visually classified in the same light conditions at the same time into groups of similar colours, described as follows: (1) dark blue/violet, (2) dark blue/black, (3) dark blue, (4) mid blue, (5) light blue, (6) grey blue, (7) greenish blue and (8) green.4.5.2.L*a*b* Colour Measurement With the L*a* b* colour coordinates, the position of a colour in the three-dimensional CIELAB Colour Space can be precisely defined.L* is the lightness axis reaching from black (L* = 0) to white (L* = 100); a* and b* are two complementary colour axes: the a* axis points toward red (positive a*) and green (negative a*), and the b* axis, toward yellow (positive b*) and blue (negative [87].L*a*b* values were determined with the Nix Quality Control Color Sensor (Nix QC) and the Nix Toolkit App both from Nix Sensor Ltd. (Hamilton, ON, Canada).To avoid contamination, each sample was measured between two cover glasses of microscope slides.

Table 1 .
Comparison of the MME and sLPE methods with both Persicaria tinctoria and Indigofera arrecta (mean values of three and two provenances, respectively); comparisons of MME and mod.MME and of sLPE and mod.sLPE with one provenance of Indigofera arrecta (mean values of extraction replications).t-test; for L*a*b*values, the aggregated data set was used.

Colour Classification Dark Blue/ Violet Dark Blue/ Black Dark Blue Mid Blue Light Blue Grey Blue Greenish Blue Green Example for colour class
2.4.2.Measured Colour Parameters (L*a*b*)

Table 3 .
Multivariate linear regression of indigotin content on L*a*b* values.The aggregated data set was used.

Table 3 .
Multivariate linear regression of indigotin content on L*a*b* values.The aggregated data set was used.

Table 4 .
Range of indigo quality and extraction yields.Minimum and maximum mean values achieved in laboratory experiments; the results from pilot or prototype extractions are indicated separately (data in italics: min/max values if extraction experiments were carried out without replications).The original data were transferred into comparable units (not included are data with other reference points, such as extraction yields referring to area (kg/ha) or percentage in weight per volume (w/v)).Only data from an indigotin analysis of indigo are reviewed; theoretical indigotin calculations from precursor analyses are not included.Only one experiment performed.(2)"?": minimum data not provided in the study.(3)Pilotscale:estimatedbasedonlaboratory-scaleresults.(4)Typin in the source quoted: "Na(OH)2" which should be either NaOH or Ca(OH)2.(5)"?": minimum data not calculated in the study when indigotin content < 2 After purification (removal of indirubin with acetone).(7)Total indigotin yield of separate two-step extraction of leaves and stems.Abbreviations: dw = d weight, fw = fresh weight,  = leaves,  = leaves and stems, n.d.= not determined, HPLC = high-performance liquid chromatography, HPTLC-densitom high-performance thin-layer chromatography-densitometry.Comparison of Strobilanthes cusia and I. tinctoria.Only with Strobilanthes cusia: test of extraction methods: different plant material conditions (fresh, semi-dried, dried), different soaking durations; extraction at ambient temperature, adding Ca(OH)2 solution, aerating, washing precipitated indigo with Ca(OH)2 solution (lab scale).

Table 5 .
Overview of the indigo sample set.

Table A3 .
Cultivation of indigo-yielding species in Vienna, Austria in 2021.Indigofera sp.: pruning of main shoot to enhance ramification and increase biomass yield, because only a few seeds were available per provenance If the leaves on the lower part of the stems had already dropped, only the upper part with leaves still attached was harvested.Indigofera species: apart from MME.45-47, MME.48-50, MME.69-71 and sLPE.28-30,very thick main stems were removed and only the thinner side branches were used for extraction.

Table A4 .
Comparison of the L*a*b* values of indigo from different species, per country.All samples extracted using the MME method (Austria: mean values of all provenances per species, China: one provenance per species; therefore, mean values of extraction replications).Values with the same letter do not differ significantly within a country (ANOVA, p = 0.05, aggregated data set).

Table A4 .
Cont.Extraction duration extended to 6 h.( 2 ) Extraction duration extended to 9 h.Mean value of two replications (the sample from one replication was too small for L*a*b* measurement).

Table A5 .
Comparison of provenances within a species: L*a*b* values of indigo samples extracted with the MME method, cultivated in Austria (mean values per provenances).Values with the same letter do not differ significantly within a species (ANOVA, p = 0.05, aggregated data set).
( 1 ) Thick stems not removed from harvested plant material.