Abstract
The effect of the modifier structure and content, as well as the dye structure, on the absorption on montmorillonite nanoparticles has been studied. The discovered dependencies have been used for obtaining nanopigments with high content of the coloring component based on cationic, direct, acid, active or disperse dyes. The effect of the coloring component on photopolymerization of the oligomer/monomer binder is estimated and the optimum compositions of ultraviolet (UV) paints with nanopigments are determined. It has been shown that developed UV paints possess higher light sensitivity compared with conventional organic pigments, and allow a dyed material with increased resistance to rubbing and tensile strength to be obtaining.
Dyed photopolymerizable oligomer/monomer compositions are widely used as ultraviolet (UV) inks in the polygraphic industry.1–3 There are a few works testifying about the prospect of UV inks’ application to pigment printing on textile materials.4–7 It is shown that optimization of the printing compositions with disperse dyes or organic pigments provides UV inks with high light sensitivity. Cotton fabrics colored by these compositions possess good resistance to physical, chemical and mechanical impacts. The disadvantage of these UV inks, especially printing compositions based on disperse dyes, is their limited range of colors, because not all dyes stand the severe conditions of color fixation. Moreover, many disperse dyes are strong photopolymerization inhibitors.7,8 At the stage of photopolymerization, organic pigments are much more stable but also cause a significant effect on this process. 8 Furthermore, their high concentrations may reduce physicomechanical characteristics of the material. All this stimulates the search for novel printing formulae, which combine advantages of disperse dyes and pigments, but are free from their disadvantages. The use of nanopigments (NPs) based on montmorillonite, which adsorbs well cationic and other dyes, is one of the important trends in this application.9–11
The aim of this study is the development of new technological approaches to UV inks production for textile materials printing using NPs.
To achieve this aim, the following basic tasks were performed:
development of the methods for obtaining NPs based on the modified montmorillonite with high dye content; analysis of the role of the modifier structure and content, as well as the dye structure on the adsorption on nanoparticles; assessment of the effect of NPs on the binder photopolymerization and their stability in this process; optimization of the UV ink composition based on NPs; determination of the morphology of binder films dyed by the NP; evaluation of physicochemical and physical and mechanical performances of colored fabrics.
Experimental details
Materials
For the main material, cotton fabric (100%, scoured bleached cotton sheeting art. 262, fabric weight 120 g/m2) produced by OJSC ‘Trekhgornaya manufaktura’ (Russia) was used. Some supplementary studies have been performed using polyamide, viscose, acetate or polyester fabrics.
Main binder components are oligourethane methacrylates – a laboratory sample based on oligotetrahydrofuran with end toluylene isocyanate groups (OUMA) with the average molecular weight of 1400 were synthesized in the Polymer Department, ICP RAS – as well as industrial products of interaction of 2,4-toluylene diisocyanate with laprol 2000 and hydroxypropylmethacrylate(OUMA-2002T) or hydroxypropylmethacrylate (OUMA MTM; ‘Khimtranzit’ LLC, Russia).
To increase light sensitivity, as a cross-linking agent trimethylol propane triacrylate (TMPTA) (‘Yarsintez’ JSC RI, Russia) has been used as received from the manufacturer.
The photoinitiators are 2,6-di-tert-butyl anthraquinone (NIOPiK, Russia) and bis-(2,4,6-trimethylbenzoyl)-phenylphosphynoxide (Irgacure 819, Ciba, Switzerland); the surfactants are polypropylene glycol (Shuchard, Germany), polyoxyethylene sorbitol monostearate (Tween 20) and polyethylene glycol (Ferak, Germany), which, along with the solvents – acetone and ethanol (Mosreaktiv CJSC, Russia) – were of analytic grade. All of the reagents were obtained from commercial suppliers without any purification.
For obtaining NPs, as coloring components dyes of various classes were used: Basic Red 18 (CI 11085), Basic Red 13 (CI 48015), Basic Yellow 24 (CI 11480), Acid Red 18 (CI 16255), Acid Violet 49 (CI 42640), Acid Yellow 76 (CI 18850), Acid Green 25 (CI 61570), Acid Red 73 (CI 27290), Acid Red 441, Mordant Red 7 (CI 18760), Direct Blue 199 (CI 74190), Direct Blue 151 (CI 24175), Direct Brown 95 (CI 30145), Direct Yellow 1 (CI 22250), Direct Red 239 (CI 29160), Direct Violet 9 (CI 27885), Disperse Blue 56 (CI 63285), Disperse Yellow 13 (CI 58900), Disperse Orange 11 (CI 60700), Disperse Red 50 (CI 11226), Disperse Blue 7 (CI 62500) produced by ‘Novocheboksarsky Chemical Plant’ OJSC, Russia, as well as Reactive Red 247 (‘Novocheboksarsky Chemical Plant’ OJSC, Russia), and Reactive Blue 38 (SCPL, India). The pigments Irgalite Blue NGA (blue pigment) and Irgalite Super Yellow SBA (yellow pigment) (Ciba, Switzerland) were used directly in the form of dry powders, as received from manufacturers. The content of the main substance in commercial samples of dyes was at least 70 wt% except disperse ones, where it was about 50–70 wt%.
Conditions of nanopigment preparation, content of dye into nanopigment and color coordinates of nanopigments
MPCS: the product of the montmorillonite modification by cationic surfactant.
Methods
Rheological characteristics of UV inks obtained by mixing components with the given contents were determined using the device Rheotest RV 2.1 (Germany).
Color coordinates of dyed samples in the CIELAB-76 system were determined on the ColorFlex spectrophotometer (HunterLab, USA). Color fastness tests for various physical and chemical factors were performed in accordance with national and international standards.12–16 Physical and mechanical characteristics of dyed fabrics were estimated according to standards.17,18
NPs' size and surface parameters of dyed compositions after photopolymerization were determined using an atomic-force microscope (NT-MDT, Russia). Small-angle X-ray scattering was carried out on a Bruker NANOSTAR (Bruker AXS Inc., USA), and electron microscopic studies were performed using a Tescan Mira 3 (Czech Republic) scanning electron microscope without additional processing of the sample surfaces. The absorption spectra in UV and visible ranges were recorded using MultiSpec-1501 (Shimadzu, Japan) and Specord UV-Vis (Germany) spectrophotometers.
Results and discussion
Basic stages of printing composition preparation and coloring of textile material
The process solution designed comprises four basic stages: (1) NP synthesis; (2) printing composition production; (3) application of a printing composition to a textile material; and (4) coloring composition fixation on a material.
NPs were synthesized by slow adding and continuous mixing of the dye solution to the product of the montmorillonite modification by cationic surfactant (MPCS) dispersion in the same solvent (alcohol, acetone or their mixture with distilled water). The amount of the dye remaining in the solution after NP separation was determined using the spectrophotometric technique. After drying and assessing the dye content (adsorbed dye), the NP obtained was used directly for production of UV paints. In some cases, to increase NP dispersibility in an oligomer/monomer composition, a surfactant (polypropylene glycol, polyoxyethylene-sorbitol monolaurate or polyethylene glycol) was added.
Table 1 shows particular examples that characterize conditions for NP preparation with both high and low content of a coloring agent (a dye). The effect of decisive factors, such as dye kind and content in solution, and modifier structure and content on the MPCS surface, is considered in the following part in more detail.
The printing compositions were made by mixing the NP and photoinitiator with a binder. For the latter, as shown before, 7 mixtures of oligomers with (meth)acrylate end groups with cross-linking agents containing two or more methacrylate or acrylate groups are desirable. To improve NP dispersion in the binder, the prepared mixture was treated using an ultrasonic disperser for 1–2 min. It is found that further increase of ultrasonic treatment duration causes no quality improvement for the printing composition and coloring on its base. In a number of instances, moreover, this may cause premature binder polymerization and, as a consequence, a sharp increase in viscosity of the printing composition.
When optimizing flow properties by varying oligomer and monomer structure and ratio, the printing composition based on oligourethane methacrylates and other oligomers may be applied on a textile material by any common method: using a screen, engraved roller or a printer. In this work, the simplest printing technique with a screen was used.
In order to fix coloring composition on the material, ready-printed fabric samples were irradiated by full light of a medium pressure Hg-lamp DRT-400. Samples were located 20 cm away from the lamp. Color quality after irradiation during a certain time interval was determined by organoleptic analysis, the presence or the absence of tackiness, and color fastness to physicochemical impacts.
When preparing and using printing compositions based on ‘common' industrial pigments, similar procedures have been applied.
Note that despite the presence of dyes applied on modified montmorillonite nanoparticles in printing compositions, these compositions should be considered exactly as the compositions with NPs. This follows from both general considerations and results of special tests testifying that all studied dyes, except for disperse ones, which have a limited solubility, are insoluble in the used oligomer/monomer binder. Thus, coarse dispersions obtained by their direct mixing with the binder are impractical for coloring textile materials.
Factors affecting content of the coloring component in nanopigment
Multiple experimental data obtained (Table 1 and Figure 1) indicate that the following three factors define the content of the coloring component (the dye) in NPs:
chemical origin and structure of the dye molecule; dye concentration and the solvent origin properties; modifying agent (surfactant) structure applied to the MPCS produced. Effect of MPCS type on quantity dyes in nanopigments: (a) Reactive Blue 38, (b) Disperse Blue 56.
For cationic dyes, chemisorption is the main process19–21 for making NPs. At relatively low dye content in the solution (≤10–15 wt% relative to MPCS), dyes are sorbed quantitatively and after NP separation, the solution is not dyed or has faint color. In case of dye content ≥20 wt%, sorption sites are saturated and the coloring component content in NP becomes independent of the dye concentration in the solution.
Dynamic equilibrium is typical for dyes of other classes: the solution obtained after NP separation is dyed even at relatively low dye concentrations. This testifies to the physical adsorption of the dye on the MPCS surface. 20
The effect on the modifier structure on the dye adsorption is rather strong: under the same conditions of the synthesis, the coloring component content in NPs may be 2–3-fold different (Figure 1). Both the scale and character of this dependence may vary with the dye structure (Figures 1(a) and (b)).
Knowing these dependencies, unfortunately, by now only on the empirical level NP may be produced with an extremely high coloring component content (up to 30–40 wt% against the initial MPCS (Table 1, Figure 1).
For the purpose of forecasting color characteristics of expected coloring by using NPs in UV inks, Table 1 also shows color coordinates of obtained NPs in dispersions in polymethyl methacrylate. Clearly, NPs give a very wide range of colors that meet applied requirements.
It is found that under the same conditions of NP production, the dye content in NPs may change several times, as the surfactant structure and the dye origin vary.
Figure 1 shows the effect of MPCS type on the dye amount applied on it. Reactive Blue 38 dye content varies broadly: from 6.3 to 19.1 wt%, whereas Disperse Blue 56 content varies from 8.7 to 19.4 wt%. However, as these dyes are applied on the same MPCS type (MPCS-7), the coloring component content in NPs is found to be rather different. The content of Reactive Blue 38 dye applied is low (6.3 wt%) (Figure 1(a)), whereas Disperse Blue 56 dye content is high enough (15.4 wt%) (Figure 1(b)).
MPCS-4 is the optimum product giving as a rule NPs with maximum content of coloring component when dyes of different classes are applied.
Data in Figure 1 and Table 1 show that dye interaction with MPCS is rather complex and depends on both the dye structures, namely, the presence of certain functional groups allowing chemical or physical bonding to the montmorillonite and the type of organo-modifier, the surfactant.
Optimization of UV ink composition based on nanopigments
As shown above, the use of OUMA as the binder for UV inks based on disperse dyes or organic pigments is preferable against its industrial analogue, Acrol U. 7 Variation of the oligomeric component in the NP-based UV inks within this work also shows some advantages of the product obtained in the laboratory as compared with its industrial analogues, OUMA-2002 T and OUMA MTM. This may be additional evidence of the principal importance of multifunctionality and purity of the oligomeric component of the UV inks. Nevertheless, as shown below, printing formulae based on ‘non-ideal’ industrial oligomers may give high-quality coloring. By use of some other oligomers – bis-FEA (etoxylated bisphenol A diglycidyl ether diacrylate), bis-FGA (bisphenol A diglycidyl ether glycerolate diacrylate) and oligoglyceryl methacrylate (OGM) – the obtained color fastness is significantly reduced. Therefore, the main part of the study was carried out with OUMA and its industrial analogues.
Effect of ultraviolet (UV) ink components on color fastness and on time of UV drying
OUMA: oligourethane methacrylate based on oligotetrahydrofuran with end toluylene isocyanate groups; bis-FEA: etoxylated bisphenol A diglycidyl ether diacrylate; bis-FGA: bisphenol A diglycidyl ether glycerolate diacrylate; TMPTA: trimethylol propane triacrylate; OGM: oligoglyceryl methacrylate.
As mentioned above, in a number of cases a surfactant is the necessary component of UV inks. This is especially important for the use of NPs with a high content of disperse dyes, when the product obtained without a surfactant represents an amorphous solid that can hardly disperse in the binder. In this case, adding a surfactant leads to a sharp increase of NP distribution in the binder.
For UV ink production, the most difficult and important task is selection of NP type and its concentration. Selection of the optimum NP content is mostly defined by the desired color and hue that, in turn, depends on the type and content of the coloring component in the NP, as well as NP combining with the binder. The latter is not a significant restriction since, as shown below, NPs with high dye content of 20–30 wt% distribute well in OUMA and its analogues. Principally more complex and an important problem is selection of NP type, which is critically dependent on the origin of dye used as a coloring component in the NP. It is clear that the NP must meet four basic requirements:
have desired and sufficiently intensive color; have minimum effect on the photopolymerization rate (UV drying time); be stable (not participate in photochemical and thermal reactions, and not be consumed) during UV drying; provide high color resistance to physical, chemical and mechanical impacts.
Shown below are multiple NP examples meeting these rigorous requirements. However, it should be taken into account that not all dyes can be used for obtaining NPs with required properties.
Thus, the optimum formulae of UV inks contains (wt%):
Basic characteristics and advantages of UV inks based on NPs
Color coordinates of cotton fabrics colored with ultraviolet (UV) inks a and of primary color standard
aOUMA-2002Т/TMPTA (69/23 wt%) binder, 2,6-di-tert-butylanthraquinone (3 wt%) as a photoinitiator.
It is quite natural to assume that adsorption of molecules of dyes in amounts comparable with surfactant content on montmorillonite may not change the nanoparticle size significantly. A presumption that in the structure of UV inks, NPs obtained are also distributed at the nanosize level rather than forming micron-sized aggregates, similar to MPCS powder, seems to be not so unambiguous. Proofs for the nanosized structure of the colored layer on material obtained using NP-based UV inks are given below.
Rheological characteristics of UV inks based on NPs differ insignificantly from those of inks with conventional pigments which, in turn, are defined by the oligomer/monomer composition properties. For instance, the flow index of UV inks with NPs based on Acid Yellow 76 dye and MPCS-4 (4 wt%) calculated from the Oswald–de Ville equation equals 0.97, whereas for the similar composition without NP or with blue pigment (2 wt%) it is 1.00 and 0.84, respectively. Thus, NP introduction into UV inks causes no effect on its viscoelastic properties, and no auxiliary complex equipment is required when operating with it.
Color fastness to physical and chemical impacts at printing by nanopigments and pigment inks a
aOUMA-2002Т/TMPTA (69/23 wt%) binder, 2,6-di-tert-butylanthraquinone (3 wt%) as a photoinitiator.
UV: ultraviolet.
Physical and mechanical characteristics of cotton fabric printing with ultraviolet (UV) inks
aOUMA-2002Т (69 wt%), TMPTA (23 wt%), 2,6-di-tert-butylanthraquinone (3 wt%), nanopigment Disperse Blue 7/MPCS-4 (5 wt%).
bOUMA-2002Т (69 wt%), TMPTA (23 wt%), 2,6-di-tert-butylanthraquinone (3 wt%), blue pigment (5 wt%).
Apparently, different effects of NP-based UV inks on the warp and weft strength is due to insufficient binder elasticity. At the warp tensile of the colored fabric, when elongation at break only equals 7%, its behavior is as a peculiar composite material. Both the fabric and the binder are destroyed almost simultaneously. At the weft tensile elongation of fabric at break is much higher, giving 15%. In this case, colored fabric is destroyed stepwise: the binder film breaks first, and then the fabric, the strength of which is much higher. These effects are also typical of a material printed by the UV inks based on a common pigment. In this case, the warp strength increases significantly as compared with uncolored fabric, but to a lesser extent than for the NP-based UV inks. The change of weft strength for printing by an ink based on a common pigment is also low.
Resistance to wearing of fabrics colored with the use of NPs is also quite high (>50,000 cycles), which significantly exceeds common requirements.
Beside the assessment results of NP-based inks color fastness, Table 4 shows minimum times of UV-light radiation (UV drying) necessary for fast color fixing. It is seen that in this aspect NPs are much better than conventional pigments. Figure 2 gives more demonstrative and detailed presentation of this. As data show, almost the same fixation rate as NPs show, common blue pigment may only achieve at abrupt, nearly by an order of magnitude, concentration decrease. This, however, leads to a significant change in the hue, which changes from dark to light.
Effect of nanopigments (1–5) and pigment (6–8) on time of ultraviolet drying: 1 – Acid Violet 49/MPCS-4 (4 wt%); 2 – Disperse Blue 7/MPCS-4 (5 wt%); 3 – Reactive Blue 38/MPCS-6 (7 wt%); 4 – Acid Yellow 76/MPCS-4 (4 wt%); 5 – Mordant Red 7/MPCS-4 (4 wt%); 6 – blue pigment (0.25 wt%); 7 – blue pigment (1 wt%); 8 – blue pigment (2 wt%).
A low coloring fixation rate in the presence of blue pigment is qualitatively recorded by an organoleptic method as the stickiness presence. The polymerization of the upper composite layer is the limiting factor for stickiness elimination. Therefore, one may suggest that deceleration of polymerization by this pigment is associated with the inhibiting action induced by the radical termination increasing on the pigment particle surface rather than with the screening one.
The reason for the absence of expressed NP retardation action is apparently polymerization proceeding in the interlayer nanosized space, where diffusion of radicals and oxygen, which is the polymerization inhibitor, is hindered.
As compared with common pigments, the abrupt decrease of UV drying time in the presence of NPs should naturally reduce consumed energy and increase productivity. This is the third, decisive, advantage of NPs as compared with common pigments.
Hence, the novelty of the developed approach with regard to textile science includes the following:
for a coloring agent of printing compositions for textile materials, NPs obtained by applying dyes on montmorillonite nanoparticles modified by cationic surfactants are suggested; it is demonstrated that rheological properties of printing compositions based on NPs and oligomer/monomer binder permit their use for coloring textile materials using traditional printing techniques, including screens; as compared with similar compositions that comprise ‘common’ organic pigments, NP-based UV inks are characterized by decreased UV drying time; fabrics printed by NP-based UV inks have improved strength; color obtained with the use of NP-based UV inks has more fastness to rubbing than color by similar inks based on ‘common’ pigments.
The nanosize structure of NPs and NP-colored compositions
One may suggest that the features and advantages of NP-based UV inks are due to their nanosize structure. Multiple data obtained with the use of a set of physicochemical methods, including optical spectroscopy, atomic-force microscopy, X-ray diffraction studies and electron microscopy, prove this thesis.
It is shown that beside the above-listed features, UV inks and films formed by them have one more feature: transparency in the visible region, beyond the absorption bands dyes used for NP preparation (Figure 3). Similar inks containing a common blue pigment, for example, are practically opaque in the visible region (in this case, at concentrations corresponding to the coloring component content in NPs, optical density is ≥3). Transparency of samples with NP evidences that NP sizes in them are not greater than the light wavelength (400–700 nm). This result correlates with our hypothesis that the main part of NPs in UV inks represents single MPCS plates with applied dye, almost uniformly distributed in the binder.
Visible spectra of photopolymerized ultraviolet inks containing oligourethane methacrylate based on oligotetrahydrofuran with end toluylene isocyanate groups, trimethylol propane triacrylate, 2,6-di-tert-butylanthraquinone and nanopigments: 1 – Acid Yellow 76/MPCS-4 (9.3 wt%), 2 – Reactive Blue 38/MPCS-4 (5.2 wt%).
However, as shown by the X-ray diffraction analysis, some part of MPCS plates with applied dye form aggregates (packs), each comprising up to several tens of plates. Figure 4 shows the investigation results for such aggregates obtained by the small-angle scattering. The presence of clearly expressed basal reflections absent in analogous films without a pigment, as well as in films with a common pigment, prove the presence of order zones provided by formation of MPCS aggregates plates (packs about 30 nm thick) with characteristic space between layers (plates) of 3.5 nm. These assessments correlate well with the data obtained for analogous samples containing MPCS in the absence of a dye (∼35 and 3.9 nm, respectively). Since the interlayer space in these aggregates exceeds the MPCS interlayer space (1.9 nm), the data obtained indicate intercalation of the oligomer/monomer binder, to almost the same extent in both NPs and MPCS without the dye. These results fully correspond to the known ideas on the structure of polymer mixtures with montmorillonite nanoparticles, in accordance with which both polymer intercalates into the interlayer space, and nanoparticles exfoliate by mixing.
22
Local microsurrounding of the dye molecules in such aggregates (oligomer/monomer binder and a surfactant fragment) does not significantly differ from their surrounding, when they are located in individual MPCS particles. As a consequence, difference in spectral characteristics should not be significant.
The small-angle X-ray scattering of photopolymerized inks containing oligourethane methacrylate based on oligotetrahydrofuran with end toluylene isocyanate groups (67 wt%), trimethylol propane triacrylate (23 wt%), 2,6-di-tert-butylanthraquinone (3 wt%): 1 – without pigment; 2 – with MPCS-4 (7 wt%); 3 – with nanopigment Reactive Blue 38/MPCS-4 (7 wt%).
Investigation of films formed from UV inks by the atomic-force microscopy method allowed one to obtain a direct proof of the presence of nanostructural formations on the coloring layer surface. Colored surface containing NP manifests quite high homogeneity. However, it has a number of protrusions 10–100 nm at the high level and 0.2–1 µm at the base (Figure 5). High heterogeneity is typical for the surface of material treated by a similar composition, but containing MPCS without a dye. It has multiple smaller hummocky protrusions with the average height of 10–40 nm and the lateral size at the base of 0.1–0.3 µm. The surface obtained with the use of compositions based on common pigment has much larger protrusions with the average height of 0.8–1.2 µm and the size at the base of 1–2 µm (Figure 6).
Atomic force microscopy image of photopolymerized inks containing oligourethane methacrylate based on oligotetrahydrofuran with end toluylene isocyanate groups (67 wt%), trimethylol propane triacrylate (23 wt%), 2,6-di-tert-butylanthraquinone (3 wt%) and nanopigment Disperse Yellow 13/MPCS-4 (7 wt%). Atomic force microscopy image of photopolymerized inks containing oligourethane methacrylate based on oligotetrahydrofuran with end toluylene isocyanate groups (71 wt%), trimethylol propane triacrylate (24 wt%), 2,6-di-tert-butylanthraquinone (3 wt%) and blue pigment (2 wt%).
Close assessments of the surface heterogeneity of the colored coating were also obtained by the method of scanning electron microscopy. It is found that the surface of the coating based on a common pigment has a microheterogeneous structure with typical size of heterogeneities of about 1–10 µm (Figure 7). Apparently, this is due to phase separation caused by a difference in polymerization rate of the oligomer and the monomer. Coating with the NP leads to a significantly different structure (Figure 8). In this case, heterogeneities of a micron size are absent, and the structure of the coating is similar to the analogous uncolored coating containing MPCS without a dye.
Scanning electron microscopy image of coating containing oligourethane methacrylate based on oligotetrahydrofuran with end toluylene isocyanate groups (69 wt%), trimethylol propane triacrylate (23 wt%), 2,6-di-tert-butylanthraquinone (3 wt%) and blue pigment (5 wt%). Scanning electron microscopy image of containing oligourethane methacrylate based on oligotetrahydrofuran with end toluylene isocyanate groups (69 wt%), trimethylol propane triacrylate (23 wt%), 2,6-di-tert-butylanthraquinone (3 wt%) and nanopigment Disperse Blue 7/MPCS-4 (5 wt%).
Conclusion
Thus, using determined dependencies of dye adsorption on modified montmorillonite nanoparticles it is possible to make NPs of a wide range of colors, with up to 30–40 wt% content of coloring component.
Printing compositions based on mixtures of NP with an oligomer/monomer binder and a photoinitiator manifest decreased UV drying time as compared with compositions based on common organic pigments.
Fabric printed by UV inks including NP have high wear resistance and improved strength and color fastness to rubbing.
The advantages of UV inks are provided by their nanosized structure proved by a set of physicochemical methods, which include spectrophotometry, X-ray diffraction analysis, atomic-force and electron microscopy.
Footnotes
Funding
This research received no specific grant from any funding agency in the public, commercial or not-for-profit sectors.