The Journal of The Textile Institute

ISSN: (Print) (Online) Journal homepage: https://www.tandfonline.com/loi/tjti20

Eco-friendly printing paste replacing urea-based

formulations in cotton printing

Susana Vílchez , Adaris López , Jonathan Miras , María José Bleda , Albert M.
Manich , Nuria Puigventós , Esteve Genís & Jordi Esquena

To cite this article: Susana Vílchez , Adaris López , Jonathan Miras , María José Bleda , Albert
M. Manich , Nuria Puigventós , Esteve Genís & Jordi Esquena (2020): Eco-friendly printing paste
replacing urea-based formulations in cotton printing, The Journal of The Textile Institute, DOI:
10.1080/00405000.2020.1797262

To link to this article: https://doi.org/10.1080/00405000.2020.1797262

Published online: 27 Jul 2020.

Submit your article to this journal

Article views: 20

View related articles

View Crossmark data

Full Terms & Conditions of access and use can be found at

https://www.tandfonline.com/action/journalInformation?journalCode=tjti20
THE JOURNAL OF THE TEXTILE INSTITUTE
https://doi.org/10.1080/00405000.2020.1797262

RESEARCH ARTICLE

Eco-friendly printing paste replacing urea-based formulations in cotton printing

peza, Jonathan Mirasa, Mar
ıa Jos
e Bledaa, Albert M. Manicha, Nuria Puigvento
Susana V
ılcheza, Adaris Lo
sb,
Esteve Gen
ıs and Jordi Esquena
b a

a
Institute of Advanced Chemistry of Catalonia (IQAC), Spanish National Research Council (CSIC), and CIBER of Bioengineering, Biomaterials
and Nanomedicine (CIBER-BBN), Barcelona, Spain; bUNITEX, Olesa de Montserrat, Barcelona, Spain

ABSTRACT ARTICLE HISTORY

In textile printing, hydrotropic components are usually added to print paste formulations promoting Received 18 September 2019
the hydration of textile fabrics by disrupting the hydrogen bonding network between water mole- Accepted 14 July 2020
cules. For this purpose, urea is commonly used but nevertheless, its presence causes water eutrophica-
KEYWORDS
tion and their reduction or replacement by more environmentally-friendly products is required. In the
Cotton; reactive dyes;
present work, we have studied an alternative hydrotropic product (polyethylene glycol, PEG) to replace urea-free printing;
urea. A factorial design of experiments has been carried out to identify the factors that optimize sev- experimental design
eral variables of interest such as viscosity of the printing paste and fabric colour strength. The results
showed that a reduction of 70 wt% of urea was achieved by using PEG 400, while preserving the per-
formance of printing formulation. The viscosity of these eco-friendly printing paste formulations was
similar to that of the original urea-based formulations and, more interestingly, fabrics showed similar
colour strength and fastness to those of fabrics printed by the conventional method.

Introduction agent in formulation of print pastes. Hydrotropes are

amphiphilic molecules, generally with a short hydrophobic
The development of environmentally friendly processes is
chain, which do not form micelles but nevertheless they
an important challenge of textile industry. These processes
increase solubility of poorly-water soluble compounds
should accomplish strict legislation aimed at reducing emis-
(Neuberg, 1916). They have been used in a wide range of
sion of contaminants. They should be more ecologically
applications, including separation processes or cleaning and
friendly at a competitive prize, by using safer dyes and
personal care formulations (Burkinshaw, 2015). In the spe-
chemicals and by reducing the cost of effluent treatment.
Nowadays, the main environmental problem lies in emission cific case of urea, it is a non-amphiphilic hydrotrope that
of waste water and its chemical load, which contains inor- promotes solubilization because its strong ability to form
ganic salts, surfactants and urea (Khatri et al., 2015). The hydrogen bonds. It is known that urea promotes protein
chemical load of waste water could be reduced by different denaturalization and increases solubility of macromolecules
methods. Replacement of the most contaminants products in water. In this context, urea is well recognized as a chaot-
by less contaminant alternatives, modification of processes ropic agent that disrupts the hydrogen bonding network of
and optimization of products could result in less release of water, and consequently, it increases the hydration of
contaminants (Khatri et al., 2015; Sheth & Musale, 2004). macromolecules.
Industrial textile processing comprises different opera- Urea is one of the most studied hydrotropes, and it is
tions such as printing. This process generates a high amount often used in textile printing and continuous dyeing proc-
of effluents which exhibits high biological and chemical oxy- esses, increasing dye solubility formulations (Burkinshaw,
gen demand, BOD and COD, respectively. Consequently, 2015). Nevertheless, the precise mechanism for enhancing
printing processes and composition of print pastes should the colouration of fibers in printing and dyeing processes
be reassessed to meet environmental requirements. There still remains unclear. Such improvement in colouration
are three main strategies to minimize effluent problems: could be associated to the role of urea in disaggregating and
using new and less contaminant technologies; treating swelling cellulose fibers, and thus, facilitating the penetra-
effectively waste water effluents and recycling process- tion of dyes. Moreover, urea promotes condensation of
ing solutions. water on printed fabrics during steaming (Brady, 1976;
Most cotton fabrics are printed using reactive dyes, Hamlin et al., 1999; Provost, 1992; Sheth & Musale, 2004).
achieving high degree of wash fastness and bright fashion Consequently, using urea in textile printing processes
colours (Xie et al., 2009). Since the introduction of these allows the desired results of colour fastness, colour yield and
dyes in 1957, urea has become essential as a hydrotropic leveling of printed textiles. Despite these advantages, when

CONTACT Susana V
ılchez susana.vilchez@iqac.csic.es Institute of Advanced Chemistry of Catalonia (IQAC), Spanish National Research Council (CSIC), and
CIBER of Bioengineering, Biomaterials and Nanomedicine (CIBER-BBN), Jordi Girona 18-26, 08034 Barcelona, Spain
ß 2020 The Textile Institute
2 S. VÍLCHEZ ET AL.

Table 1. Compositions of print paste formulations following the randomized in printing formulations were anhydrous potassium carbon-
factorial 32 design matrix, and definition of the three experimental levels for
ate, supplied by Probus, SeraV print M-ALV10, which is a
R

factors x1 (urea concentration) and x2 (PEG 400 concentration).

natural thickener, supplied by DyStar, LyoprintV RTA is a
R

Coded levela Experimental composition

synthetic thickener provided by Archroma, and LudigolV is
R

Experiment number x1 x2 x1 (urea, wt%) x2 (PEG 400, wt%) an oxidizing agent. Urea was supplied by Indukem, and
1 0 0 3 3.5 polyethylene glycol 400 (PEG 400) by Sigma-Aldrich.
2 1 0 6 3.5
3 1 1 6 6
4 1 1 0 1
5 0 1 3 1 Experimental design
6 0 1 3 6
7 1 1 0 6 A 32 factorial design was performed to conduct a compre-
8 1 1 6 1 hensive study of the two factors, urea and PEG 400 concen-
9 1 0 0 3.5
10 0 0 3 3.5 trations. A preliminary study was performed to determine
a
Coded levels: low (-1); central (0); and high (þ1). the suitable range of the different factors. These previous
results allowed to select a range of PEG concentrations
prints containing residual urea are washed-off, decompos- between 1 and 6 wt%, while reducing the urea concentration
ition in aqueous effluents produces nitrogenous compounds to values between 0 and 6 wt%. It was expected that urea
that seriously deteriorate water environments. Because of could be decreased from 12.5 wt%, which was the concentra-
that, urea reduction or elimination is one of the major chal- tion currently used in the industrial process. The selected
lenges in textile printing. three experimental levels of coded and uncoded factors are
The amount of urea in the printing pastes has already shown in Table 1. Nine elemental experiments configured
been reduced applying different procedures (Khatri et al., the matrix of the design, and a total of 10 experiments
2015). However, this reduction has proven not enough to were carried out after adding an extra central point.
avoid high levels of Nitrogen concentration in effluent Three variables of interest (responses) were optimized in
released waters, which causes water eutrophication. For this the factorial design of experiments: viscosity, colour
strength (K/S) and colour difference. Only models with
reason, the drastic reduction of urea concentration in print
statistically significant parameters (a ¼ 0.05) were chosen
formulations is of utmost importance, becoming a great
to obtain the best adjusted polynomial equations, and
challenge for textile industries in recent years. Different
contour graphics were drawn. The adjusted determination
approaches for elimination or replacement of urea in print-
coefficient (R2 adj) accounts for the variability of the varia-
ing have been described to deal with this problem (Khatri et
bles explained by second order polynomial terms of fac-
al., 2015; Ahmed et al., 2015). Moisture application, partial
tors, including their interactions. Non-significant terms
or complete substitution of urea by alternative products
were excluded from equation of the models. The statis-
(Sheth & Musale, 2004; Zhang et al., 2016), replacement of
tical analysis was done using Statgraphics 18 statistical
sodium alginate by synthetic thickener and developing new
software (StatgraphicsV, 2017).
R

reactive dyes (Ahmed et al., 2006) are some of

the strategies.
In view of the fact that environmental regulations regard- Printing
ing stream pollution are becoming stricter, the replacement
of urea for alternative products is essential. In the present Printing pastes contained the following ingredients, for a
study, a more environmental friendly product, polyethylene- total of 1000 g.
glycol (PEG), has been studied as a substitute for urea in
print-paste formulations. A factorial design of experiments  Dye: 42.5 g
has been carried out to find the best print conditions while  Urea: 0  11.25 g
using the minimum urea concentration. Several variables of  PEG: 3.7  22.5 g
interest have been optimized: viscosity of printing paste, col-  Natural thickening agent: 40 g
 Synthetic thickening agent: 14.45 g
our strength and colour difference. Colour fastness to water,
 K2CO3: 15 g
perspiration and rubbing were also evaluated.
 Oxidizing agent: 9.02 g
 Water: 832.0  875.3 g
Experimental
Fabrics were printed by the screen printing technique,
Materials
and then steamed at 101.5  C for 20 min. Finally, fabrics
The fabrics were woven cotton/elastan blends 97/3 with were washed with a nonionic surfactant aqueous solution at
190 g/m2 of mass per unit area. Two printing formulations 95  C for 10 min, and rinsed with water.
were used, with either red or turquoise blue colours. The
reactive dyes were Procion red PX-4B, which is a mono-
Determination of viscosity
chlorotriazine reactive dye with azo chromophore, and
Novacron PGR 50 turquoise, that is a monochlorotriazine The viscosity of the print pastes was measured with a rhe-
reactive dye with ftalocianine chromophore. Other products ometer AR-G2 (T.A. Instruments). A flow continuous ramp
 THE JOURNAL OF THE TEXTILE INSTITUTE 3

Fastness testing
The colour fastness to washing, perspiration and rubbing of
printed samples were evaluated according to UNE-EN ISO
105-E01:1996, UNE-EN ISO 105-E04:1996 and UNE-EN
ISO 105-X12:1996, respectively.

Results and discussion

Cotton is a highly crystalline cellulose fibre, in which stiff
crystalline regions consists of densely packed chains.
Cellulose is a linear polysaccharide consisting of D-glucose
units linked together by b(1!4) glucosidic bonds. It is well
known that these chains tend to remain in the solid state,
because of intra- and inter-molecular hydrogen bonds, lead-
ing to an organized crystalline structure (Brant & Goebel,
1972). Dye molecules cannot penetrate the highly ordered
Figure 1. Surface tension of solutions at different PEG 400 concentrations, with
and without urea. The control sample was the original formulation, with
crystalline regions of native cellulose and their adsorption
12.5 wt% urea, in absence of PEG 400. can only take place inside micropores of amorphous regions
(Beltrame et al., 1998).
was performed. Measurements were performed at 25  C In textile printing, urea acts as a solvation aid, allowing
with 40 mm plate-plate geometry and a gap of 1000 mm. the swelling of cotton fibers and promoting a faster penetra-
Flow curve determinations were carried out after applying a tion of dye. Polymer swelling is defined as a process where
pre-shear of 30 s1 for 1 min, and using a shear rate from solvent molecules penetrate into the polymer structure to a
0.01 to 500 s1. All measurements were made by duplicate. certain extent, leaving the volume and physical properties of
the biopolymer significantly altered while the solid, or semi-
solid fractions, remain practically unchanged (Medronho &
Colour strength measurement Lindman, 2015). For the reactive dye, urea performs as a
water-absorbing agent in the pre-treatment of print pastes,
Kubelka-Munk values (K/S) of printed samples were meas-
increasing fibre hydration during steaming process, and
ured at either kmax ¼ 520 nm, in the case of red dye, or thus, urea accelerates migration of dye from the thickener
680 nm when using turquoise blue dye (Equation 1). A spec- film, i.e. sodium alginate, into cotton fibres (Yuen
trophotometer Macbeth Colour-Eye 3000 (Macbeth, USA) et al., 2004).
with D65 illuminant and 10 observer was used. K/S values The current work describes the use of an eco-friendly
were calculated following equation 1. potential substitute of urea, PEG 400, able to form hydrogen

bonds with cellulose chains, and thus, enhancing cellulose
K
¼ ½ð1RÞ2 =2R (1)
S swelling and improving dye penetration into cotton fibers.
where R is the reflectance; K is the adsorption coefficient; In addition, this urea substitute should show at least similar
and S is the scattering coefficient. The higher K/S value, the hydrotropic properties than urea. As it was mentioned
above, urea is used in textile printing and continuous dyeing
better colour strength.
processes to increase dye penetration because of its hydro-
Colour difference (DE) between two samples was calcu-
trope properties (Burkinshaw, 2015). Often, urea is also con-
lated by means of L, a and b colour coordinates
sidered a chaotrope since it weakens the H-bond structure
(Equation 2). L indicates luminosity, a is the red/green
of water (De Xammar Oro, 2001). Then, first of all, the
coordinate and b is the yellow/blue coordinate.
potential substitute of urea should behave similarly. Surface
qffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi tension of water can be used to evaluate the activity of a
DE ¼ ðDL Þ2 þ ðDa Þ2 þ ðDb Þ2 (2) chaotrope, since these agents reduce surface tension, as
opposite to kosmotropes, which increase surface tension.
The lower DE value, the more similar to control sample. Therefore, surface tension of various aqueous solutions of
For the three optimized variables, measurements were urea and PEG 400 was measured. The results, expressed as
done by duplicate and the mean value was used. the mean of four replicates (Figure 1), could reflect the
main role of PEG 400 in printing pastes. By increasing PEG
Surface tension measurements 400 concentration from 1 to 12.5 wt% without urea, surface
tension tends to decrease, which could reflect the chaotropic
Surface tensions measurements of selected urea and PEG behavior of PEG 400. This is observed when comparing to
400 solutions were measured at 25  C in a Sigma 700 tensi- values of surface tension in presence of urea. The addition
ometer (KSV Instruments Ltd.) by the Wilhelmy plate of 6 wt% of urea only produces a small additional reduction
method. Four replicates were performed and the mean value in surface tension, indicating that PEG 400 might be as
was calculated, with standard deviation values below 3%. effective as urea in reducing tension.
4 S. VÍLCHEZ ET AL.

Figure 2. Viscosity plot of print paste corresponding to Procion red (a) and Novacron turquoise (b) control formulations.

Figure 3. Viscosity contour surface levels, as a function of urea and PEG 400 concentrations, (a) formulations with Procion red; (b) with Novacron turquoise ( iden-
tifies optimal regions).

In any case, it should be pointed out that PEG 400 might tension from 60.3 mN/m (at 12.5 wt% urea without PEG
have a higher degree of amphiphilicity than urea. 400) to 55.1 mN/m (with 12.5 wt% PEG in absence of urea).
Consequently, PEG 400 can contribute more than urea to Consequently, the incorporation of PEG 400 could enhance
the decrease in surface tension. PEG 400 might reduce ten- the penetration of dye into fabrics by a reduction of the sur-
sion by two different mechanisms: acting as a chaotropic face tension. To sum up, all these results clearly indicate
cosolvent and also by adsorbing on water surface because of that PEG 400 could perform the role of hydrotrope, simi-
a certain degree of amphiphilicity. larly to urea, hopefully with improved performance.
The content of urea of the print paste control sample Viscosity is a crucial parameter that should be controlled
was 12.5 wt%. PEG 400 produces a decrease of surface to obtain a uniform printed fabric. All print pastes showed
 THE JOURNAL OF THE TEXTILE INSTITUTE 5

Table 2. Results for red and turquoise dyes, showing the optimal polynomial equations of regression models for viscosity, by urea
(x1) and PEG 400 (x2) concentration factors.
Viscosity
Procion red 9.84  3.61x1 þ 4.43x12 þ 2.30x2 – 2.24x1x2 þ 4,97x1x22 – 1.28x12x22
R2adj ¼ 94.5%
Novacron turquoise 11.18 þ 0.55x1 þ1.9525x12 þ 1.47x2 þ3.57x22 – 1.12 x12x2 – 3.09 x12x22
R2adj ¼ 90.7%
R2adj: Adjusted determination coefficient of the regression model.

Table 3. K/S and DE of fabrics printed with either Procion red or Novacron Regarding the results achieved with Novacron turquoise,
turquoise, at different compositions of urea and PEG 400. viscosity values show a similar trend (Figure 3b). Minimum
Procion red Novacron turquoise and maximum values are obtained with 3 wt% urea, with
Experiment number K/S DE K/S DE both 3.5 wt% and 6 wt% PEG 400 (experiments 10 and 6),
1 21.1 1.88 12.7 1.37 respectively. Again, the viscosity closest to that of control is
2 22.8 0.52 12.50 1.38 achieved at the minimum PEG 400 concentration (experi-
3 21.6 1.24 11.8 1.31 ment 10). The same trend is observed by using both dyes. It
4 23.2 0.60 12.5 0.62
5 22.7 0.20 14.3 0.85 confirms that PEG 400 concentration is the factor that most
6 21.6 1.50 11.5 2.23 influences viscosity. This is the expected result, since PEG is
7 22.4 0.86 11.0 3.03
8 23.1 0.93 15.5 3.51
a polymer that might play the role of a thickener.
9 22.3 0.77 11.2 2.75 From the experimental results, a multiple regression ana-
10 21.7 1.85 11.6 2.38 lysis was carried out to find the optimal fitted equation.
Control 23.0 13.4
Only the statistically significant parameters were selected
(Kleinbaum et al., 1988; Mitchell, 2012; Pe~ na, 2002). This
procedure was performed for both Procion red and
a pseudoplastic or shear-thinning behavior with an initial
Novacron turquoise (Table 2) in presence of urea (x1) and/
plateau, at low shear stress, in which viscosity is independ-
or PEG 400 (x2). Goodness of model fitting was evaluated
ent of the shear stress applied. Above a certain shear rate
and adjusted determination coefficients (R2adj) above 90%
value, viscosity decreases linearly due to the rupture of the
were obtained.
inner structure of the system. As an example, viscosity of The contour surface plots for both dyes are shown in
control formulations by using Procion red and Novacron Figure 3. Different levels of viscosity are shown as function
turquoise is shown in Figure 2. Print pastes should show of both urea and PEG 400 concentrations. The minimization
shear-thinning under shear and restoration of original vis- of viscosity was the desired target because a value as similar
cosity after the removal of shear, obtaining sharp and clear as possible to control sample is required.
drawing patterns (Wang et al., 2013). The contour plots for Procion red (Figure 3a) and for
Any eco-friendly substituent of urea should provide tar- Novacron turquoise (Figure 3b) show the minimum values
get values of viscosity (similar to those of control samples) of viscosity, indicated in the whitest areas, at 6.0  7.5 and
to achieve a suitable application of print pastes on fabrics. 11.0  12.0 Pas, respectively. The results obtained for
The study of its rheological behavior is essential as it is rec- Procion red demonstrate that the minimum value of 7.5 Pa
ognized that rheological properties play a very important may be achieved between 0 to 3 wt% of urea and between 1
role in textile printing. The print paste viscosities of control to 1.25 wt% of PEG 400. This means that these concentra-
samples (12.5 wt% urea without PEG 400) with Procion red tions of urea and PEG 400 provide the same rheological
and Novacron turquoise were 7.5 Pas and 10.7 Pas, respect- properties than the control sample, with 12.5 wt% of urea in
ively. Viscosity values obtained from the experimental absence of PEG 400. Therefore, 1 wt% PEG 400 can allow
design with Procion red ranges between 6.9 and 17.9 Pas. an almost complete substitution of urea.
On the other hand, with Novacron turquoise, values vary in Using Novacron turquoise, the minimum viscosity value
the same range approximately, between 11.1 and 16.2 Pas. (11.0 Pas), represented as the white spot (Figure 3b), is
Both dyes show viscosity values close to that of control sam- obtained with a composition between 2.4 and 2.7 wt% urea
ples, meaning that new formulations present similar rheo- with 3 wt% PEG 400. It should be remarked that these
logical properties. results represent the achievement of 80% decrease in urea
Accordingly, results point to different optimal composi- concentration, in comparison to the current compositions of
tions, depending on the dye. Using Procion red, the min- print pastes.
imum viscosity, 6.9 Pas, is obtained in experiment n 4 (0 In summary, the results shown in Figures 3 confirm, as
wt% urea, 1 wt% PEG 400). In addition, this viscosity value expected, that it is possible to substitute a large percentage
is very close to that of the control sample (7.5 Pas). On the of urea with an eco-friendly alternative, without affecting
contrary, experiment n 9 (0 wt% urea, 3.5 wt% PEG 400) viscosity. As mentioned before, the control of viscosity in
shows the highest value, 17.9 Pas. This could indicate that print pastes is crucial to obtain good results of fastness and
PEG concentration is the variable that most influences vis- colour difference.
cosity, as expected. Concentrations of urea up to 6 wt% The next step was evaluating the final properties of
show intermediate viscosity values between 10 and 14 Pas. printed fabrics, comparing their colour properties values
6 S. VÍLCHEZ ET AL.

Table 4. Optimal polynomial equations of the regression models for K/S and DE, by urea (x1) and PEG 400 (x2) concentration factors. Fabrics were printed with
either Procion red or Novacron turquoise.
Procion red Novacron turquoise
K/S 21.53 þ 1.07x12 – 0.61x2 þ 0.63x22 – 0.24x1x22  0.64x12x22 11.93 þ 0.84x1 – 0.46x2 – 0.53x1x2  0.89x12x2 þ 0.74x12x22
R2adj ¼ 85.47% R2adj ¼ 93.40%
DE 1.84  0.17 1  1.20 12 þ 0.65 2  0.96 22  0.52x12x2 þ 0.34x1x22 þ 1.97  0.71 1 þ 0.65 2  0.42 22  1.17x1x2  0.58x12x2 þ
1.27x12x22 0.98x1x22 þ 0.60x12x22
R adj ¼ 99.02%
2
R adj ¼ 87.99%
2

R2adj: Adjusted determination coefficient of the regression model.

Figure 4. K/S (a) and DE (b) contour surface levels of fabrics printed with
Procion red, as a function of urea and PEG concentrations ( identifies opti-
mal regions). Figure 5. K/S (a) and DE (b) contour surface levels of fabrics printed with
Novacron turquoise ( identifies optimal regions).

with fabrics printed using the control composition Table 5. Colour fastness to washing, perspiration and to rubbing of fabrics
(12.5 wt% urea without PEG 400). First of all, colour printed with Procion red, for wool (W) and cotton (CO) fabrics.
strength (K/S) and colour difference (DE) were determined Washing Acid perspiration Basic perspiration Rubbing
Experiment
by spectrophotometry. These results are shown in Table 3. number W CO W CO W CO Dry Wet
The optimal compositions would be those producing 1 4/5 4/5 4 3/4 4/5 4 4/5 3
maximum K/S values (that indicate highest colour yield) 2 4/5 4/5 3/4 4 4 4 4/5 2/3
3 4/5 4/5 4 3/4 4/5 3/4 4/5 3
and minimum DE (which means minimal colour difference, 4 4 4 3/4 4 4 4 4/5 2/3
in comparison to control). Colour differences could only be 5 4 4 3/4 4 4 4 4/5 2/3
acceptable with DE values below 1. It should be noted that 6 4 4 3/4 4 4 4 4/5 2/3
7 4/5 4 4/5 4 4/5 3/4 4/5 3
K/S not only increases due to improved fixation of dye on 8 4/5 3/4 4 3/4 4/5 3/4 4/5 2/3
fabrics, but it is also influenced by lack of penetration (asso- 9 4/5 4 4/5 4 4/5 3/4 4/5 3
ciated with higher viscosity of print pastes) and higher 10 4/5 4 4/5 4 4/5 4 4/5 2/3
Control 4 4 4 4 4 4 4/5 2/3
amounts of applied print pastes, which can be associated to
a lower viscosity (Kumbasar & Bide, 2000).
We are interested in maximizing K/S and minimizing 5) also presents a high value of K/S (22.7), similar to that of
DE. Comparing values as a function of percentage of urea control sample, and the lowest value of DE (0.2).
and PEG 400 for Procion red dye, the maximum K/S value On the other hand, the use of Novacron turquoise shows
of 23.2 is obtained in the experiment n 4, which corre- a slightly different behavior. The maximum K/S value corre-
sponds to 0 wt% urea and 1 wt% PEG 400. Moreover, this sponds to experiment n 8 (6 wt% urea, 1 wt% PEG 400).
shows a relatively low value of DE (0.6). In addition, the Nevertheless, this composition shows the highest value of
composition 3 wt% urea and 1 wt% PEG 400 (Experiment n DE (3.51), indicating that it is not a proper formulation.
 THE JOURNAL OF THE TEXTILE INSTITUTE 7

The lowest DE (0.62) is shown by experiment n 4 (0 wt% the results illustrate that it could be possible to minimize
urea, 1 wt% PEG 400), which provided a K/S value (12.5) DE by mixing 1.25 wt% PEG 400 to urea between 0.6 and
not far from control (13.4). Experiment n 5 also shows an 4.6 wt%, which reflects the same tendency obtained with K/
acceptable K/S value (14.3), however its DE (0.85) is higher S. Hence, these results indicate that incorporation of the
than that of experiment n 4, but the aim was minimizing PEG 400 gives rise to a positive effect on the colour strength
DE. However, these results illustrate that it is possible to of fabrics, with low PEG 400 concentrations. Indeed, such
formulate print pastes, almost without urea. It should be eco-friendly product allows reaching an important decrease
remarked that we have obtained printed fabrics with K/S in urea concentration, removing around 90%.
values similar to control (13.4, obtained with 12.5 wt% The contour plots obtained for Novacron turquoise are
urea), while minimizing DE (< 0.62). shown in Figure 5. Overall, differences can be observed
Following the same procedure than for evaluation of vis- between this dye and Procion red (Figure 4), comparing val-
cosity, the same statistical analysis was carried out for K/S ues of maximum K/S and minimum DE. In the case of
and DE, to find the optimal fitted equations for Procion red Novacron turquoise, results indicate that it is possible to
and Novacron turquoise (Table 4), as a function of urea (x1) maximize K/S values, achieving values of 15.0, higher than
and PEG 400 (x2) concentrations. control (13.4) with a composition of 6 wt% urea and 1 wt%
The adjusted determination coefficient (R2adj) could be PEG 400 (Figure 5a). Concerning DE parameter (Figure 5b),
considered acceptable with values higher than 85% (Draper values of 0.8 can be obtained using urea concentrations
& Smith, 1981). Using these fitted equations, the contour between 0 and 3.4 wt% and 1.0  1.25 wt% PEG 400 and
surface plots for K/S and DE responses were obtained also at 6 wt% urea and 3.75  5.6 wt% PEG 400. Then, being
(Figures 4 and 5), achieving optimal composition with the conservative and fixing an urea concentration of 6 wt%, we
lowest urea concentration. have demonstrated that it is possible to reduce urea concen-
Analysing the contour surface levels obtained with tration by approximately 50% (from 12.5 wt% urea of con-
Procion red (Figure 4a), it is clearly shown that K/S values trol composition) with Novacron turquoise.
are higher with a composition in the range between 1 and Colour fastness is one of the most relevant criteria for
2.25 wt% of PEG 400 and 0 to 1.2 wt% of urea, meaning an quality printing, which is also a measure of the stability of
important reduction of urea (almost 90%), while preserving the dye-fibre linkage. The colour fastness to washing, per-
K/S values similar to those of control sample ( 23.0). spiration (acid and basic) and rubbing (dry and wet) of
Regarding the colour difference (DE) of fabrics (Figure 4b), printed fabrics were evaluated to determine the suitability of
these compositions, compared to those of the control. The
Table 6. Colour fastness to washing, perspiration and to rubbing of fabrics values are summarized in Tables 5 and 6 for Procion red
printed with Novacron turquoise, for wool (W) and cotton (CO) fabrics. and Novacron turquoise, respectively. Regarding fabric stiff-
Washing Acid perspiration Basic perspiration Rubbing ness, large variations were not observed during the prelim-
Experiment
number W CO W CO W CO Dry Wet inary tests and consequently, this property was not
1 3/4 3 4 3 3 3/4 4/5 4 measured and it was excluded from the multifactorial
2 3/4 3 3/4 3 3/4 3/4 4/5 3/4 experimental design.
3 3/4 3 3/4 3 3 3 4/5 3/4
4 4 3 4 3/4 4 3/4 4/5 3/4
As seen in Table 5, all analysed compositions show
5 4/5 3/4 4 3/4 4 3/4 4/5 3/4 acceptable values, comparable to those of control. Most of
6 4/5 4 4/5 4 4/5 4 4/5 3/4 the printed fabrics did not exhibit important changes in
7 4/5 4/5 4/5 4/5 4/5 4/5 4/5 4
8 2 2 2 2/3 2 2/3 4/5 4/5 their visual aspect. Only a slight staining on the adjacent
9 4/5 4/5 4/5 4/5 4/5 4/5 4/5 4 fabric, in the case of wet rubbing, was observed (values of
10 4/5 4/5 4/5 4/5 4/5 4/5 4/5 4 colour fastness between 2 and 3). Moreover, these values
Control 4/5 3/4 4/5 3/4 4/5 4 4/5 3/4
were similar to those of the control. Such results indicate

Figure 6. Desirability contour plots approaching the target values for viscosity, K/S and DE, as a function of urea and PEG 400 concentrations, for fabrics printed
with Procion red (a) and Novacron turquoise (b) formulations ( identifies optimal regions).
8 S. VÍLCHEZ ET AL.

that many studied formulations could replace the current The two dyes show a different behavior when maximiz-
composition used in print pastes (control), achieving an ing the desirability function, by jointly optimizing viscosity,
important reduction in urea concentration. K/S and DE. In print pastes with Procion red (Figure 6a),
When printing pastes are prepared with Novacron tur- the maximum D (0.8) could be reached with 6 wt% urea
quoise (Table 6), colour fastness values are slightly lower and 2.6  4.4 wt% PEG 400 (as a function of experimental
than those obtained with Procion red. As it was discussed factors). Nevertheless, another suitable desirability value of
above, this result could be related to lower viscosity of print 0.7 could be reached with an important reduction of urea.
pastes with Procion red, in combination together with lesser Such region is located on the bottom of the plot, where a
fixation of Novacron turquoise dye into cotton fabrics maximum in the desirability function is observed at 1 wt%
(Kumbasar & Bide, 2000). In this system, the compositions PEG 400 concentration, with 2.4  3.9 wt% urea. This result
that provided higher values of colour fastness were experi- reflects that it could be possible to reduce urea concentra-
ments 6, 7 and 9 which corresponds to 0.0  3.0 wt% urea tion with a removal between 69 to 81 wt%.
and 3.5  6.0 wt% PEG 400. These results on colour fastness When Novacron turquoise dye is used in print pastes
demonstrate that it is possible to greatly reduce urea con- (Figure 6b), the maximum desirability (0.6) is found in the
centration, without damaging the colour yield of fabrics region comprised between 0 and 2.7 wt% of urea and 1 and
while maintaining similar properties to current dye 1.75 wt% of PEG 400. In summary, it should be highlighted
formulations. that the incorporation of a small percentage of PEG 400 in
In order to optimize together the three variables of inter- printing pastes (< 1.5 wt%) seems to allow an important
est (responses), viscosity, K/S and DE, the desirability func- reduction of urea concentration (above 78%). Moreover,
tion was used (Del Castillo et al., 1996; Montgomery, 2001). these printed fabrics show quite similar viscosity and colour
The general approach is first to convert each response yi strength properties, compared with those obtained with the
into an individual desirability function di that varies over control composition.
the range 0 di 1, where if the response yi is at its goal In order to confirm these results, cotton fabrics were
target, then di ¼ 1, and if the response is outside an accept- printed with the optimal printing paste, the composition of
able region, di tends to zero. In this current work, the pur- which was obtained from the maximization of the desirabil-
sued target was equaling the viscosity of control composition ity function. Hence, when printing with Procion red a con-
(7.5 Pas for Procion red and 10.7 Pas for Novacron tur- centration of 3.15 wt% of urea and 1 wt% of the PEG 400
quoise, Figures 2a and 2b, respectively), maximizing K/S and was used. In the case of Novacron turquoise, the optimal
minimizing DE. The closer to the pursued targets, the greater concentrations were 1.35 wt% urea and 1.5 wt% PEG 400.
individual desirability approaching to 1. The further to tar- These experiments were performed by duplicate.
gets, the lower desirability approaching to zero. Tables 7 and 8 show the results, obtained for these opti-
The optimization of multiple responses is based on the mal concentrations, of viscosity, K/S, DE and colour fast-
maximization of the whole desirability, D, given by the geo- ness, respectively. As it can be seen, K/S values are similar
metric mean of the three considered individual desirabilities to those obtained with the control composition and DE is
d1 (desirability of viscosity), d2 (desirability smaller than 1. Moreover, good results of colour fastness
pffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi of K/S) and d3 were obtained.
(desirability of DE), being D ¼ 3 d1 d2 d3 . The contour
surface plots for this whole desirability D function are pre-
sented in Figure 6 for Procion red (a) and Novacron tur- Conclusions
quoise (b).
This study has provided a quantitative insight, using a stat-
Table 7. Print paste viscosity, K/S and DE of fabrics printed with Procion red istical analysis approach, into the incorporation of an eco-
and Novacron turquoise at optimal concentrations. friendly product as a potential substitute of urea in printing
Procion red Novacron turquoise pastes, with reactive dyes in cotton fabrics. The results have
Replicates Viscosity (Pas) K/S DE Viscosity (Pas) K/S DE clearly demonstrated the suitability of polyethylene glycol
Optimal 1 11.5 22.7 0.3 16.3 13.6 0.7 400 to partially replace urea. First of all, surface tension
2 11.4 21.8 0.5 13.6 12.5 1.0 measurements have shown a certain amphiphilic nature of
Control 1 7.2 22.5 _ 11.2 14.5 _
2 7.8 23.0 _ 10.3 13.4 _ PEG 400 that could indicate improved hydrotropic proper-
ties, in comparison with urea. The use of the desirability

Table 8. Colour fastness to washing, perspiration and to rubbing of fabrics printed with Procion red and Novacron turquoise (W: wool, CO: cot-
ton) at optimal concentrations.
Washing Acid perspiration Basic perspiration Rubbing
Experiment W CO W CO W CO Dry Wet
Optimal red 4 3 3/4 3 3/4 3 4/5 2/3
Optimal red replicate 4 3 3/4 3 3/4 3 4/5 2/3
Control red 4 3 3/4 3 3/4 3 4/5 2/3
Optimal turquoise 3 1/2 3 1/2 2/3 1/2 4/5 3
Optimal turquoise replicate 3 2 3/4 2 3/4 2/3 4/5 3
Control turquoise 3 1/2 3/4 2 3 1/2 4/5 3/4
 THE JOURNAL OF THE TEXTILE INSTITUTE 9

functions has enabled to simultaneously optimize viscosity, Burkinshaw, S. M. (2015). The role of water in aqueous dyeing
K/S and DE, showing optimal responses with reductions of (Chapter 3). In Andrew Filarowski (Ed.), Physico-chemical aspects of
urea above 70%. This important reduction in urea concen- textile coloration (pp. 81–124). Wiley. https://doi.org/10.1002/
9781118725658
tration did not produce any decrease of colour fastness. The De Xammar Oro, J. R. (2001). Role of co-solute in biomolecular stabil-
achieved results can be comparable, or even better, than ity: Glucose, urea and the water structure. Journal of Biological
those obtained using the control composition. Physics, 27(1), 73–79. https://doi.org/10.1023/A:1011890506834
These results are promising and they could demonstrate Del Castillo, E., Montgomery, D. C., & McCarville, D. R. (1996).
that it is possible to greatly reduce urea concentration, by Modified desirability functions for multiple response optimization.
using an eco-friendly substitute of urea, while preserving Journal of Quality Technology, 28(3), 337–345. https://doi.org/10.
1080/00224065.1996.11979684
properties similar to currently used compositions. That Draper, N. R., & Smith, H. (1981). Applied regression analysis (2nd
would constitute an important reduction in wastewater ed.). John Wiley & Sons.
treatments, providing a less contaminant solution for print Hamlin, J. D., Phillips, D. A. S., & Whiting, A. (1999). UV/visible spec-
paste formulators. troscopic studies of the effects of common salt and urea upon react-
ive dye solutions. Dyes and Pigments, 41(1-2), 137–142. https://doi.
org/10.1016/S0143-7208(98)00076-X
Acknowledgements Khatri, A., Peerzada, M. H., Mohsin, M., & White, M. (2015). A review
on developments in dyeing cotton fabrics with reactive dyes for
The authors are very grateful to the European Regional Development reducing effluent pollution. Journal of Cleaner Production, 87,
Fund, which provides support to government programs. Viscosity 50–57. https://doi.org/10.1016/j.jclepro.2014.09.017
determination of print pastes has been performed by the Kleinbaum, D. G., Kupper, L. L., & Muller, K. E. (1988). Applied
Nanostructured Liquid Characterization Unit, located at the Institute regression analysis and other multivariable methods (2nd ed.).
of Advanced Chemistry of Catalonia (IQAC), belonging to the Spanish
Boston: PWS-Kent Publishing Company.
National Research Council (CSIC) and affiliated to the NANBIOSIS
Kumbasar, E., & Bide, M. (2000). Reactive dye printing with mixed
ICTS of the Biomedical Networking Center (CIBER-BBN).
thickeners on viscose. Dyes and Pigments, 47(1/2), 189–199. https://
doi.org/10.1016/S0143-7208(00)00075-9
Medronho, B., & Lindman, B. (2015). Brief overview on cellulose dis-
Disclosure statement solution/regeneration interactions and mechanisms. Advances in
No potential conflict of interest was reported by the author(s). Colloid and Interface Science, 222, 502–508. https://doi.org/10.1016/j.
cis.2014.05.004
Mitchell, M. N. (2012). Interpreting and visualizing regression models
Funding using Stata. Stata Press.
Montgomery, D. C. (2001). Design and analysis of experiments. John
The authors acknowledge financial support from EEA-Grants Program Wiley. https://doi.org/10.1002/qre.458
(grant no. 20140916). The authors also greatly acknowledge the finan- Neuberg, C. (1916). Hydrotropy. Biochem Z, 76, 107–176.
cial support from the Spanish Ministry of Economy and Pe~
na, D. (2002). Regresi
on y dise~no de experimentos. Alianza Editorial.
Competitiveness (grant no. CTQ2017-84998-P) and support from Provost, J. R. (1992). Effluents improvement by source reduction of
Generalitat de Catalunya (grant nos. 2017SGR1778 and TECCIT15- chemicals used in textile printing. Journal of the Society of Dyers
1-0009). and Colourists, 108(5-6), 260–264. https://doi.org/10.1111/j.1478-
4408.1992.tb01456.x
Sheth, G. N., & Musale, A. A. (2004). Substitute Products for Urea in
Application of Reactive Dyes to Cotton Fabrics. Indian Journal of
References Fiber & Textile Research, 29, 462–469.
StatgraphicsVR Centurion 18 (2017). Statgraphics Technologies, Inc.
Ahmed, I., Ahsan, M. L., Silva, A., & Armenta, S. (2015). Reducing Wang, L., Zhu, F., Yang, Q., & Lu, D. (2013). Rheological properties of
effluent pollution in dyeing cotton fabrics with reactive dyes.
modified xanthan and their influence on printing performances on
Spanish Journal of Industrial Chemistry and Textile Physics, 6,
cotton with reactive dyes in screen printing. Cellulose, 20(4),
50–57.
2125–2135. https://doi.org/10.1007/s10570-013-9968-5
Ahmed, N. S. E., Youssef, Y. A., El-Shishtawy, R. M., & Mousa, A. A.
Xie, K., Liu, H., & Wang, X. (2009). Surface modification of cellulose
(2006). Urea/alkali-free printing of cotton with reactive dyes.
with triazine derivative to improve printability with reactive dyes.
Coloration Technology, 122(6), 324–328. https://doi.org/10.1111/j.
1478-4408.2006.00047.x Carbohydrate Polymers, 78(3), 538–542. https://doi.org/10.1016/j.
Beltrame, P. L., Castelli, A., Selli, E., Mossa, A., Testa, G., Bonfatti, carbpol.2009.05.013
A. M., & Seves, A. (1998). Dyeing of cotton in supercritical carbon Yuen, C. W. M., Ku, S. K. A., Choi, P. S. R., & Kan, C. W. (2004).
dioxide. Dyes and Pigments, 39(4), 335–340. https://doi.org/10.1016/ Study of the factors influencing colour yield of an ink-jet printed
S0143-7208(98)00016-3 cotton fabric. Coloration Technology, 120(6), 320–325. https://doi.
Brady, P. R. (1976). The effect of urea on wool fibers. Journal of the org/10.1111/j.1478-4408.2004.tb00238.x
Society of Dyers and Colourists, 92(2), 56–58. https://doi.org/10. Zhang, H., Gao, A., Song, X., & Hou, A. (2016). Cleaner production
1111/j.1478-4408.1976.tb03275.x applied to urea-free printing of cotton fabrics using polyethylene
Brant, D. A., & Goebel, K. D. (1972). Evidence for the occurrence of glycol polymers as alternative additives. Journal of Cleaner
flexible sugar ring conformers in cellulosic chains. Macromolecules, Production, 124, 126–131. https://doi.org/10.1016/j.jclepro.2016.02.
5(4), 536–538. https://doi.org/10.1021/ma60028a035 090