Lightfastness of polypropylene pigmented with natural indigo of Isatis Tinctoria

Owing to increasing eco-awareness, natural alternatives to synthetic colorants are in demand (Křížová, 2015; Niinimäki, 2022; Pattanaik et al., 2020; Samanta and Agarwal, 2009; Shahd and Mohammad, 2013). Natural colorants, which are colorants derived for instance from plants, fungi or minerals, answer this need by offering a more environmentally friendly alternative to petroleum-based synthetic colorants (Kaddoura et al., 2019; Nambela et al., 2020; Panda et al., 2022; Räisänen, 2009; Yusuf et al., 2017). Natural colorants can be completely different from synthetic colorants, such as tannins, or the same substances as synthetic colorants with natural production pathways, such as indigotin (Habib et al., 2022; Miyoko and Ryoko, 2009). Compared to synthetic colorants, natural colorants generally use fewer toxic chemicals and are less harmful to aqueous environments (Ali et al., 2009; Cristea and Vilarem, 2006; Lu et al., 2020; Rehman et al., 2018). Moreover, many natural colorants are biodegradable and nontoxic to humans, whereas most synthetic colorants are not (Ali et al., 2009; Cristea and Vilarem, 2006; Rehman et al., 2018; Samanta and Agarwal, 2009). However, many natural colorants reportedly have duller shades and worse fastness properties than synthetic colorants (Clarke and Anliker, 1980; Nambela et al., 2020; Rehman et al., 2018). In addition, many natural colorants do not have a constant composition owing to variance in raw material properties, which leads to variance in the properties of the extracted colorant (Merdan et al., 2017; Nambela et al., 2020).

Plant-based compounds that are commonly used as natural colorants include carotenoids, flavonoids, quinones and indigo (Garcia-Macias and John, 2004; Khoo et al., 2017; Merdan et al., 2017; Yusuf et al., 2017). Indigo is an abundant synthetic vat dye used to color jeans; however, it can also be traditionally acquired via hot water extraction from the leaves of various plant species (Blackburn et al., 2009; John, 2009; Travis, 2007). In temperate climates, one of the preferred species for natural indigo production is the cold-resistant Isatis Tinctoria, commonly known as woad or dyer’s woad (Baran et al., 2010; John and Angelini, 2009; Orsini et al., 2012). Although water-soluble precursors of natural indigo from Isatis Tinctoria are traditionally used to dye clothes, they can also be made into a stable indigo blue pigment by oxidizing and drying the extract from indigo plants (Blackburn et al., 2009; Maugard et al., 2001; Mocquard et al., 2022; Novotná et al., 2003). The indigo pigment can be used in coatings, such as paint, or extruded into thermoplastics to produce a deep blue color (Jordan et al., 2022; Jordan and Laaksonen, 2023; Mocquard et al., 2022). Mixing natural indigo with thermoplastics is an interesting topic in some published studies (Jordan and Laaksonen, 2023; Van Den Oever et al., 2004; Velho et al., 2017).

One of the most used thermoplastic polymers is polypropylene, which is found in nonwoven fabrics, packaging and automotives (Dahlbo et al., 2018; Maddah, 2016). Main advantages of polypropylene are its well-rounded mechanical properties and low market price (Kissel et al., 2004; Roes et al., 2007). Recent research has shown that polypropylene is also suitable for additive manufacturing, which is currently a common method for manufacturing highly customized and complex products (Chikkangoudar et al., 2021a; Christakopoulos et al., 2022a; He and Lv, 2023; Zander et al., 2019). However, the semi-crystallinity of polypropylene can cause unwanted shrinking and warping during thermal processing (Bachhar et al., 2020; Christakopoulos et al., 2022b; Karian, 2003). Therefore, finding optimal printing conditions and polymer characteristics for the additive manufacturing of polypropylene is a topic still under development (Bachhar et al., 2020; Chikkangoudar et al., 2021b; Christakopoulos et al., 2022b).

To obtain durable products, high lightfastness is required for many applications (Aydemir and Yenidoğan, 2018; Kim and Park, 2001). In contrast to many other natural colorants, the lightfastness is a widely known advantage of natural indigo (Crews, 1987; Cristea and Vilarem, 2006; Padfield and Landi, 1966; Seixas De Melo et al., 2004). However, previous research has shown that the lightfastness of commercial natural indigo from small-scale agricultural production does not always match the lightfastness of synthetic indigo, presumably because of impurities in natural indigo (Jordan and Laaksonen, 2023). Various purification processes have been tried in the past, for instance, dissolving the pigment in chloroform and drying the solution (Chanayath et al., 2002; Tayade and Adivarekar, 2014). Furthermore, an unpublished final report from Spindigo project suggests washing the pigment powder with HCl (Spindigo, 2004).

To further investigate the effect of purification on lightfastness of natural indigo, a commercial woad-based natural indigo pigment was purified with HCl. HCl was chosen as the purifying agent due to its ability to remove metallic impurities from the pigment without using complex extraction methods or hazardous organic solvents. Moreover, unlike weaker acids such as acetic or citric acid, HCl can easily dissolve any possible plant matter residue amongst the pigment. Thermogravimetric analysis (TGA) and Fourier-transform infrared (FTIR) spectroscopy were used to determine effects of HCl purification on the pigment composition. Furthermore, the lightfastness properties of unpurified natural indigo, HCl-purified natural indigo and synthetic indigo were investigated in a solid polypropylene (PP) matrix and in a dimethyl sulfoxide (DMSO) solution. Finally, the lightfastness properties of multiple indigo-colored 3D-printed polypropylene samples were investigated by applying xenon arc light. The test results were assessed using a reflectance spectrophotometer.

Natural indigo pigment was provided by Natural Indigo Finland Ltd and synthetic indigo powder manufactured by Sigma-Aldrich Ltd was obtained from VWR International Ltd. Polypropylene (PP) powder for compounding was obtained from Goonvean Fibres Ltd. DMSO in ≥99.9% purity for dissolving indigo pigments was obtained from Aquarius Pro Life Ltd and hydrochloric acid in ≥99% purity for HCl purification of natural indigo was obtained from WVR International Ltd.

Natural and synthetic indigo pigments were ground into a fine powder using a planetary ball mill (Fritsch Pulverisette 6 with zirconium oxide grinding balls). Grinding was done in ten one-minute periods at a rotational speed of 550 rpm. After grinding, 10 g of natural indigo powder was mixed with 2 M HCl in a beaker and the mixture was kept at room temperature for 24 h. Distilled water was then added to the HCl-pigment mixture and the pigment was separated from the liquid by centrifugation. The water addition and centrifugation processes were repeated four times. The washed pigment was then dried at 50°C for 24 h and ground with an IKA A 10 Basic batch mill, resulting in 5,115 g of fine HCl-purified indigo powder.

To determine the relative metallic impurity content of the pigments, TGA of purified natural indigo, unpurified natural indigo and synthetic indigo pigments were performed using a thermogravimetric analyzer (Pyris 1 TGA). A small amount of each pigment was heated from 30°C to 700°C at a rate of 10°C/min, and the sample weight was monitored throughout heating. After heating, the relative amount of metallic impurities was determined by comparing the percentage weight loss of each pigment.

FTIR spectral analysis with attenuated total reflection (ATR) was used to detect possible differences between pigment compositions. The ATR-IR transmittance spectra of the pigment powders in the wavenumber range of 400–4000 cm⁻¹ were measured using a Shimadzu IRSpirit-X FTIR spectrometer.

To test the absorbance and lightfastness of each indigo pigment in a DMSO solution, 0.1 mg of each pigment was dissolved in 20 ml of DMSO. Pigment solutions were exposed to xenon arc light in a xenon test chamber (Heraeus Suntest CPS+) for 5, 15 and 30 min. Before and after each exposure period, the light absorbances of the solutions in the wavelength range of 260–700 nm were measured using a UV-Vis spectrophotometer.

The indigo pigments were compounded with PP powder at four different compositions. In addition, an unpigmented reference was made in the same way. Plastic filaments were formed by the extrusion of PP powder at a temperature of 200°C using a twin-screw extruder. The molten plastic was cooled on a belt conveyor and collected as continuous monofilaments. Compositions of the filaments are presented in Table 1.

After extrusion, each filamented sample was 3D-printed into three 50-mm long, 10-mm wide and 2-mm thick sample rods using an Artillery Sidewinder X2 tabletop 3D-printer. A polypropylene printing base was used to prevent warping and sample dislocation during printing. After 3D-printing, the sample rods were exposed to continuous xenon arc light to simulate effect of indoor light exposure. The light exposure procedure was carried out in a Heraeus Suntest CPS+ xenon test chamber according to the ISO 105-B02:2014 standard, exposure cycle A1. During the lightfastness test, the color parameters and reflectance spectra of three parallel samples were measured using a Datacolor 600 reflectance spectrophotometer after 0, 24, 72 and 168 h of exposure.

The results of TGA of the indigo pigments are presented in Figure 1. The unpurified natural indigo pigment exhibited the lowest weight loss, retaining 32% of its original weight at 700°C. A substantially higher weight loss was observed for the HCl-purified natural indigo sample, of which only 21% was left after the temperature reached 700°C. The higher weight loss for HCl-purified natural indigo suggests that purification successfully removed some of the metallic impurities in the natural indigo pigment. However, only 2% of the synthetic indigo was left after the temperature had risen to 700°C, suggesting that synthetic indigo contained significantly fewer metallic impurities than its natural counterpart even after HCl purification.

Figure 1 also shows that the natural indigo pigments started to gradually decompose below 390°C, which is the theoretical sublimation point of indigo. In contrast, synthetic indigo started rapidly decomposing only when the temperature approached 390°C. The gradual decomposition of natural indigo pigments at low temperatures suggests that apart from metallic impurities, other impurities may have been left in the pigment matrix after the pigment was extracted from Isatis leaves.

IR spectra of all three pigment powders were measured to further investigate the differences between the indigo pigments. As shown in Figure 2, all pigments exhibit the typical IR peaks of indigo. For instance, the carbonyl group of indigo can be identified around 1600 cm⁻¹ and the ring structures are represented at 876, 1061 and 1167 cm⁻¹ as previously seen in a study by Baran et al. (2010). As also presented by Baran et al. (2010), Figure 2 shows that the IR peaks of natural indigo from Isatis Tinctoria are substantially less intense than those of synthetic indigo (Baran et al., 2010). The lower intensity of the natural indigo IR peaks is likely due to metallic impurities from the natural indigo extraction process. For example, calcium carbonate is often used in natural indigo extraction, and it has a wide IR peak extending from 1000 to 2500 cm⁻¹, potentially interfering with the typical IR peaks of indigo (Nooh et al., 2016; Saraya and Rokbaa, 2016).

Extending previous observations, Figure 2 also shows that the IR peaks of HCl-purified natural indigo are in the same wavenumbers as unpurified natural indigo peaks, but substantially more intense. The increased intensity of the IR peaks suggests that HCl purification decreases the amount of interfering impurities but does not substantially affect the properties of the indigo. However, despite purification, the IR peaks of the HCl-purified indigo were not as intense as the synthetic indigo peaks. Moreover, Figure 2 also shows that IR peaks of both natural indigo pigments are slightly shifted towards higher wavenumbers, which could also be a sign of interference from metallic impurities. However, HCl purification did not affect this subtle shifting phenomenon.

To further examine effect of HCl purification on the lightfastness and visual properties of indigo, the absorbance of each pigment solution was measured after 0, 5, 15 and 30 min of exposure to xenon light. Figure 3(a) shows the absorbance spectra of the solutions before and after the exposure. First, Figure 3(a) shows that before exposure, the absorbance curves of each solution had a similar shape. This shape, typical for dissolved indigo, was previously observed by Novotná et al. in 2003 (Novotná et al., 2003). However, the HCl-purified natural indigo solution had substantially higher absorbance than the unpurified natural indigo solution but much lower absorbance than the synthetic indigo solution. This indicates a higher amount of colorant in the HCl-purified natural indigo than in the unpurified natural indigo. Notably, the peak intensity for HCl-purified indigo is almost three times the intensity of the unpurified indigo, even though the HCl purification only removed approximately 51% of the dry weight of the unpurified pigment. After xenon light exposure, the absorbance at the peak maxima decreased substantially, suggesting that the dissolved pigment degraded during light exposure.

Figure 3(b) illustrates the behavior of the absorbance intensity of the 620 nm peak during exposure to each pigment solution. As shown in Figure 3(b), the synthetic indigo solution retained its absorbance proportionally better over time than both natural indigo solutions, especially during the first 15 min of exposure. This indicates a lower degradation rate of the synthetic indigo pigment in solution than that of natural indigo. Furthermore, as shown in Figure 3(b), the HCl-purified natural indigo solution retained its absorbance at 620 nm in the same way as the unpurified natural indigo solution. This implies that even though the unpurified natural indigo solution contains more colorant, HCl purification had no substantial impact on the degradation rate of the natural indigo pigment.

Photographs of the 3D-printed samples before and after the lightfastness test are shown in Figure 4. As shown in Figure 4, all the pigments produced a smooth and even initial color. Visually, each sample containing natural indigo had similar deep blue initial color, even although the unpurified sample with 0.25% natural indigo was initially a bit lighter than the other two samples. In contrast, the synthetic indigo sample was substantially darker than the natural indigo samples, despite containing equal or lower amounts of pigment. The pure polypropylene sample was colorless and partly translucent, and therefore had a minimal effect on the hue of the pigmented samples. Finally, none of the pigments caused negative effects, such as deformation or warping, during the 3D-printing process.

As described in the ISO 105-B02 standard, lightfastness ratings from 1 (very poor) to 8 (excellent) were assessed based on visual examination of the exposed and unexposed PP samples. During the xenon light exposure, each sample containing natural indigo pigment underwent a visible change towards a lighter color. Nevertheless, the samples with 0.25% HCl-purified natural indigo and 0.50% unpurified natural indigo both obtained a good lightfastness rating of 5.0. In comparison, more substantial fading was observed in the sample with 0.25% unpurified natural indigo, which only achieved a moderate lightfastness rating of 3.0. In contrast to the natural indigo samples, the synthetic indigo sample underwent very subtle color changes, resulting to and excellent lightfastness rating of 7.0. The polypropylene sample did not change visibly, fully maintaining its colorless and translucent appearance, and thus acquired a perfect lightfastness rating of 8.0. The visually assessed lightfastness ratings for the PP samples are presented in the Table 2.

Before, during and after xenon light exposure, reflectance in the wavelength range of 360–700 nm and CIELab color parameters L* (lightness), a* (color position in the green-red axis) and b* (color position in the blue-yellow axis) of the polypropylene samples were measured to track reflectivity changes in the samples. The color parameters were further used to calculate CIEDE2000 total color difference ΔE₀₀ according to procedure described by Sharma et al. (2005). Additionally, K/S value indicating color intensity of the plastics was determined from reflectance at λ = 620 nm, which is the most absorptive wavelength of indigo pigments in visible range. The K/S values were calculated according to equation (1):

The average reflective properties calculated from three parallel specimens of each plastic during light exposure are presented in Figure 5. The Figure 5(a) shows that lightness of all colored plastics substantially increased during the exposure period. Interestingly, lightness of the plastic with 0.25% of unpurified pigment increased the most despite having the highest initial lightness amongst the colored samples. However, the differences in lightness changes between the natural indigo samples may not bear statistical relevance. Nevertheless, the increase in lightness was substantially more minimal for the synthetic indigo samples compared to all natural indigo samples.

The a* and b* color parameters over exposure time are presented in Figures 5B and C, respectively. Notably, as seen in Figure 5(b), the initial color of the synthetic indigo sample was substantially redder compared to its natural counterparts. Furthermore, the plastic with 0.25% of unpurified pigment showed substantially greener initial color than the other natural indigo samples. During the light exposure, the plastic with HCl-purified indigo and shifted toward more intense blue color, as shown in Figure 5(c). Similar to the sample with HCl-purified pigment, both samples with unpurified pigment shifted toward blue until 72 h of exposure. However, unlike the plastic with HCl-purified pigment, the plastics with unpurified pigment started reverting back to less intense blue color after 72 h of exposure.

The total color changes ΔE₀₀ calculated from the CIE color parameters are presented in Figure 5(d). Among the natural indigo pigmented samples, the plastic with 0.25% of unpurified pigment showed larger color change after 168 h of exposure compared to the other two plastics. Unlike the plastics with natural indigo pigment, the unpigmented and synthetic indigo pigmented plastics both presented total color changes smaller than 2.3, which is often regarded as a threshold for visually detectable color difference.

The total reflectance in 360–700 nm wavelength range and the color intensity values based on reflectance at 620 nm over time are shown in Figures 5(e) and (f), . Figure 5(e) shows that the plastic with 0.25% unpurified natural indigo had substantially higher initial reflectance and also presented larger increase in reflectance over time compared to other plastics with natural pigment. Furthermore, as seen in Figure 5(f), it presented a substantially lower initial color intensity as well. However, all natural indigo pigmented plastics experienced a similar loss in color intensity over exposure time. Expectedly, no substantial changes in reflectance or color intensity were detected in either unpigmented or synthetic indigo pigmented plastics.

Overall, Figures 5(a–f) show that the plastic colored with HCl-purified indigo had more intense color and presented fewer changes over time in lightness, total color and reflectance compared to the plastic with equal amount of unpurified natural indigo. However, all color parameters and reflectance of the synthetic indigo sample were substantially more stable than those of the HCl-purified natural indigo sample. These differences suggest that HCl-purified natural indigo has substantially better lightfastness than unpurified indigo, but at the same time substantially worse lightfastness than synthetic indigo in PP matrix. Interestingly, the reflective properties of the plastics with HCl-purified indigo and 0.50% of unpurified indigo were remarkably similar, suggesting that lack of purification may be compensated by using more unpurified pigment. Finally, neither color parameters nor reflectance of the unpigmented PP changed substantially during light exposure, which is a sign of the excellent lightfastness of the polypropylene itself. Owing to the excellent lightfastness of polypropylene, it is highly likely that the polypropylene matrix had no effect on the lightfastness results achieved with any of the pigments.

A comparison of HCl-purified and unpurified natural indigo pigments suggests that HCl purification successfully decreased the impurity content of the pigment. After TGA, non-decomposed remains of HCl-purified pigment weighed 34% less than non-purified pigment, indicating at least 34% less impurities in the purified pigment. In addition, the increased intensity of the IR peaks suggests a lower impurity content in the HCl-purified pigment. However, based on the very low residue after TGA and the most pronounced IR peaks, it is apparent that synthetic indigo contained substantially fewer impurities than HCl-purified natural indigo. Thus, it must be noted that a substantial amount of impurities remained in the natural indigo pigment even after HCl purification.

The increased purity had a positive effect on the coloration properties and lightfastness of the natural indigo pigment. A spectrophotometric analysis showed that HCl purification substantially increased absorbance of the pigment in DMSO solution, suggesting improved coloration properties. Better coloration properties were observed in the polypropylene matrix as well, where the HCl-purified natural indigo pigment had not only a darker color but also substantially better lightfastness than its unpurified counterpart. This makes HCl purification an effective method for increasing the lightfastness and color intensity of natural indigo. However, with the least impurities, the synthetic indigo exhibited superior coloration and lightfastness properties in the polypropylene matrix.

This study showed that both unpurified and HCl-purified indigo pigments compounded well with polypropylene powder and produced a smooth and even color. As expected, light exposure only affects the pigment and not polypropylene, which makes it possible to compare the lightfastness performance of the different pigments in the polypropylene medium. Decreasing the amount of impurities in the natural indigo pigment substantially improved its lightfastness properties, making it more appealing for practical use. However, the lightfastness performance of natural indigo was worse than that of synthetic indigo even after HCl purification. This observation, along with the results from TGA and FTIR spectroscopy, suggest that a substantial amount of impurities remained in the natural indigo pigment even after purification. Based on these results, HCl purification is highly recommended to improve pigment purity in natural indigo production. However, further research on natural indigo purification is warranted, as other purification methods must be applied alongside HCl purification to achieve a high-purity product comparable to synthetic indigo. To develop effective removal processes for rest of the impurities, for instance, determining exact chemical compositions of the unpurified and purified pigments is recommended. Finally, studying more environmentally friendly alternatives to HCl is recommended to maximize sustainability of the purification process.

The authors are grateful for Mishra Rashmi for her help and guidance in the extrusion process.