Research Article pubs.acs.org/journal/ascecg
Green Finishing of Cotton Fabrics Using a Xylitol-Extended Citric Acid Cross-linking System on a Pilot Scale Jian Liu,†,‡ Bijia Wang,†,‡ Xiaomei Xu,†,‡ Jiangang Chen,†,‡ Luyi Chen,†,‡ and Yiqi Yang*,§,∥ †
Key Laboratory of Science and Technology of Eco-Textiles, Ministry of Education, and ‡College of Chemistry, Chemical Engineering and Biotechnology, Donghua University, 2999 North Renmin Road, Shanghai 201620, China § Department of Textiles, Merchandising and Fashion Design, and ∥Department of Biological Systems Engineering and Nebraska Center for Materials and Nanoscience, University of Nebraska-Lincoln, HECO Building, Lincoln, Nebraska 68583-0802, United States S Supporting Information *
ABSTRACT: Cross-linking is frequently applied to cotton fabrics for enhanced wrinkle recovery and dimensional stability. The combination of citric acid (CA) and xylitol shows great potential as a sustainable alternative to the marketdominating N-methylol resins, which are inherently formaldehyde-releasing. This paper reports a successful pilot-scale application of this green cross-linking system preceded by systematic investigation using response surface methodology (RSM). Responses of fabric properties to five foremost variables were investigated to gain insight of the cross-linking system and facilitate its industrialization. The model obtained by RSM suggests that curing temperature is the most prominent variable and the responses to CA and xylitol concentrations are closely coupled. The optimum conditions used for the pilot-scale experiments were 3 min, 175 °C, 130 g/L, 15 g/L, and 3 kg/cm2 for curing time, curing temperature, CA concentration, xylitol concentration, and padder-roll pressure, respectively. The CA/xylitol finished fabrics were comparable to those finished with the market-dominating dimethyloldihydroxyethyleneurea (DMDHEU) resins. Analyses show that CA/xylitol is more cost-effective than other formaldehyde-free cross-linking agents and clearly has a more preferable environmental, health, and safety (EHS) profile than DMDHEU. The encouraging results indicate that CA/xylitol has great potential in replacing Nmethylol resins on an industrial scale. KEYWORDS: Cotton cross-linking, Pilot scale, Sustainable, Response surface methodology, Citric acid
■
INTRODUCTION Cross-linking is frequently used in chemical processing of cellulosic textiles to overcome their inherent drawbacks of poor wrinkle recovery properties and dimensional instability.1 As the demand for easy-to-launder and wrinkle-resistant cotton fabrics continues to dominate the market, the consumption of the cross-linking agents, also known as durable press (DP) finishing resins, grows steadily.2 According to an investigative report, the consumption of durable-finishing resins has been growing at an average rate of 8.02% in the past five years. The total consumption volume of DP resins is predicted to reach 18 kilotons by 2019.3 The commercial resins found on the market are primarily produced from methylolation of amino containing substances, such as urea, cyclic ureas, carbamates, amides, and aminotriazines with formaldehyde. 4 The dominant product, dimethyloldihydroxyethyleneurea (DMDHEU), is essentially an equilibrating mixture of formaldehyde, glyoxal, and urea. A considerable amount of formaldehyde could be released from fabrics finished with DMDHEU.5 With increasing awareness of environmental concerns and ever more stringent legislation on © XXXX American Chemical Society
limitation of confirmed hazardous chemicals, DMDHEU had been partially replaced with its etherified derivatives,6 which could reduce the levels of fabric-released formaldehyde to below 100 ppm, especially when combined with thorough afterwashing.7 However, formaldehyde emissions during resin production, storage, and application still pose a threat to the environment and human health.8,9 Our test results showed that six commercial low-formaldehyde DP resin samples contained 3000 to 6000 ppm of free formaldehyde.10 Poly(carboxylic acid)s (PCAs) are widely accepted as the most promising formaldehyde-free cross-linkers for DP finishing of cellulosic textiles and treatment of other materials.11,12 However, they have very limited commercial success. PCA-based resins account for less than 10% of the formaldehyde-free DP resin market share, and less than 1% of the whole DP resin market. The dominating formaldehyde-free resin, dimethyl dihydroxyethyleneurea (DMeDHEU), resemReceived: October 2, 2015 Revised: December 5, 2015
A
DOI: 10.1021/acssuschemeng.5b01213 ACS Sustainable Chem. Eng. XXXX, XXX, XXX−XXX
Research Article
ACS Sustainable Chemistry & Engineering
Figure 1. Chemical structures of representative cross-linking agents for cellulosic textiles.
environmental, health, and safety (EHS) attributes of the CA/ xylitol system were also evaluated against that of DMDHEU. In addition, the cost-effectiveness of the CA/xylitol system was compared to commercially available DP resins of the formaldehyde-free and low-formaldehyde types.
bles DMDHEU in structure (structures shown in Figure 1), so that the well-developed finishing procedures can be easily transferrable. In contrast, the PCA cross-linking system is considerably different than DMDHEU and could be incompatible with prior finishes. For example, sodium hypophosphite (SHP), the most commonly used catalyst for PCA finishing, had been reported to cause discoloration in reactively dyed fabrics.13 Moreover, the highly acidic finishing conditions were found to be destructive for certain fluorescent whitening agents, causing severe fabric yellowing.14 Although yellowing may not be a big concern in applications as wood binders,15 severe discoloration is not acceptable in textile processing. The other challenge for the commercialization of PCA-based DP finishing agents is the overall cost performance. 1,2,3,4-butanetetracarboxylic acid (BTCA), the most effective PCA cross-linker, is over ten times more expensive than DMDHEU,16 while the economically competitive citric acid (CA) suffers from less favorable performance, durability, and more severe yellowing.17,18 Our group recently reported that the addition of polyol extenders effectively makes the CA cross-linking system more efficient by rendering it self-cross-linkable.19,20 At the same time, discoloration is mitigated upon formation of the more thermally stable polyol citrates. One of the most effective extenders identified was xylitol. From a sustainable point of view, the CA/xylitol cross-linking system has a clear advantage because both CA and xylitol are readily generated from renewable raw materials.21 The CA/xylitol cross-linking system also has great potential as an economical and green binder in replace of DMDHEU and other N-methyloyl resins for cellulosic materials22,23 in general. In this paper, we have brought the sustainable CA/xylitol cross-linking system a step further to industrialization by demonstrating its successful pilot-scale applications on cotton fabrics. To bridge the gap between lab-scale and multikilometer-scale finishing experiments, response surface methodology (RSM) was used to systematically investigate the responses of fabric properties to process parameters. The three most relevant properties of the finished fabric, namely wrinkle recovery angle (WRA), tear strength (TS), and whiteness (WI), as well as normalized-and-weighted-sum (NWS) of all three of them were evaluated as responses against curing temperature, curing time, citric acid concentration, xylitol concentration, and pressure of the padder-roll. Pilot-scale finishing of two different types of cotton fabrics was subsequently carried out based on the RSM findings. The
■
EXPERIMENTAL SECTION
Materials. For lab-scale finishing experiments, a bleached cotton poplin fabric (135 g/m2) was used. For pilot-scale finishing experiments, a bleached jacquard cotton fabric (154 g/m2) and a navy graph check cotton fabric (116 g/m2) were used. All three fabrics were manufactured by Luthai Textile Co., Ltd. CA, SHP, sodium hydroxide, hydrochloric acid, anhydrous sodium sulfite, thymolphthalein, and magnesium chloride were purchased from Sinopharm Chemical Reagent Co., Ltd. Xylitol was purchased from Adamas-Beta Reagent. CA, SHP, xylitol, and BTCA were industrial grade and the rest of the chemicals were reagent grade. All chemicals were used as received. Commercial DP finishing resins were obtained from corresponding venders as product samples. F-ECO was a low-formaldehyde resin that consists mainly of alcohol-modified DMDHEU. NF and WFF were formaldehyde-free DP resins of the imidazolidinone type. SRD787 was a commercial poly(carboxylic acid) type DP resin. Wetting agent JFC was from Wacker Chemical Company. The siloxane based softener Siligen-SIO was from BASF. Laundry detergent was a nonbleaching product purchased from Shanghai White Cat Co., Ltd. Laboratory-Scale Finishing Experiments. Preweighed fabrics (15 g) were soaked in an appropriate finishing formula and padded with a laboratory-scale Rapid M-TENTER. Two dips and two nips were applied and the fabrics were subsequently dried (100 °C, 1 min) and cured using a DaeLim Starlet DL-2015 curing machine. The cured fabrics were machine washed and tumble dried using a XQG50-1 washer−dryer according to the American Association of Textile Chemists and Colorists (AATCC) Testing Method 124 before testing. For CA/xylitol finishing, the formula compositions, padder-roll pressure, and curing conditions were varied according to the RSM experimental design. Finishing experiments with commercial DP resins and industrial grade BTCA were carried out under conditions listed in Table 1. The formulas and curing conditions of the four commercial
Table 1. Formula and Curing Conditions for Finishing with Commercial DP Resins and BTCA formula 1 2 3 4 5 B
curing conditions
F-ECO: 240 g/L, MgCl2: 24 g/L NF: 120 g/L, MgCl2: 40 g/L WFF: 80 g/L, MgCl2: 30 g/L SRD787: 160 g/L, SHP: 80 g/L BTCA: 85 g/L, SHP: 40 g/L
150 175 150 160 175
°C, °C, °C, °C, °C,
3 2 3 3 3
min min min min min
DOI: 10.1021/acssuschemeng.5b01213 ACS Sustainable Chem. Eng. XXXX, XXX, XXX−XXX
Research Article
ACS Sustainable Chemistry & Engineering
a = 1, b = 0.5, and c = 4. The levels of the five coded independent variables are given in Table 2.
resins were recommended by the manufacturers for best finishing results. The formula and curing condition for BTCA were determined based on literature reports24,25 and preoptimization experiments conducted in our lab. Pilot-Scale Experiments. Pilot-scale finishing experiments were conducted at Luthai Textile Co., Ltd. The fabrics were padded and predried on a Monfongs-828-ES TwinAir tenter frame. The roller pressure of the tenter frame was set at 1.2 bar, and the fabrics were fed at a speed of 30 m/min, which allowed the padded fabrics to dry over the steam heaters in 1.5 min. Predrying temperature was set to be 100 °C, with online temperature monitor readings ranging from 95 to 105 °C. A cylinder dryer manufactured by Kyoto Machinery Co., Ltd. was used to cure the padded fabrics. The fabrics were fed to the curing machine at a speed of 50 m/min, which converts to a curing time of 3 min. The cured fabrics were consequently sanforized on a Monfongtex-868 sanforizer. For CA/xylitol finishing, the curing temperature was set at 170 °C, and the formula contained 130 g/L citric acid, 15 g/L xylitol, 88 g/L SHP, 2 g/L JFC, and 20 g/L softener. For F-ECO finishing, curing temperature was set at 145 °C, and the formula contained 240 g/L resin, 24 g/L MgCl2, 2 g/L JFC, and 20 g/ L softener. The set curing temperatures in pilot-scale experiments were 5 °C lower than that in the lab experiments. This was because the real fabric temperatures in the lab curing machine were 5−7 °C lower than the set values while the difference between actual and set values was within 2 °C for the industrial cylinder dryer. Fabrics Tests. WRA of finished fabric was measured using Shirley crease recovery testers according to AATCC Testing Method 66-2003. CIE WI was tested on a Datacolor-650 spectrophotometer according to AATCC Testing Methods 11-2005. Tear strength of the treated fabrics was measured using a Thwing-Albert Elmendorf tearing tester according to the American Society of Testing Materials (ASTM) Testing Method D-1424-1996. All mechanical tests were carried out in the warp directions. Samples of the pilot-scale finishing experiments were sent to the Intertek Group for testing. TS was measured according to GB/T 3917.1 using a Thwing-Albert Elmendorf tearing tester. Break strength (BS) was measured according to GB/T 3923.1 using a H10K−S Tiniius Olsen elongation tester. For BS measurements, samples were cut into 50 mm strips. Color difference (CD) was tested according to GB/T 8424.2 using Datacolor-650 spectrophotometer. DP rating and dimensional stability (DS) were measured according to GB/T 13769 and GB/T 8630 respectively. Prior to the tests, the samples were subjected to a five-circle wash-and-dry sequence using a Wascator front-loading-horizontal-drum Washing machine and a Wascator Tumble drier. Wash load ballast of 2 kg, ECE Reference Detergent (0.077%), sodium perborate (0.02%), and TAED (0.003%) were used in the washing process, and the washing temperature was set to 50 °C. The amount of free formaldehyde in commercial DP resins was determined according to GB/T 5543-2006. The method was based on the reaction of formaldehyde and sodium sulfite at the pH of 9.3−10.5 to form 1 equiv of sodium hydroxylmethanesulfonate and sodium hydroxide. The sodium hydroxide generated in the process was titrated using thymolphthalein as the indicator. RSM Design. Design Expert 8.0.6 Trial (State Ease, Inc., Minneapolis, MN, USA) was used for RSM design. A five-level-fivefactor central composite design (CCD) was applied in this work. The method is suitable to optimize the final results with a minimum number of experiments, as well as to research the interactions of various parameters.26 Curing temperature, curing time, CA concentration, xylitol concentration, and pressure of the padder-roll were the five independent experimental parameters. Three measurable properties of the treated cotton fabric, namely, WRA, WI, and TS, as well as the NWS of all three of them were studied as responses. NWS was calculated using the following equation
Table 2. Coded and Uncoded Levels of Variables for RSM CCD Design coded factor levels variable
symbol
−2
−1
curing time (T, min) curing Temperature (temp, °C) citric acid concentration (CA, g/L) xylitol concentration (xyl, g/L) pressure (P, kg/cm2)
A B
1 130
2 145
3 160
4 175
5 190
C
75
100
125
150
175
D
5
12.5
20
27.5
35
E
1
2
3
4
5
0
1
2
Generally, the CCD is comprised of 2n factorial runs, 2n axial runs, and nc central runs, where n is the number of controllable variables.27 The center points nc are used to evaluate the error and reproducibility of experiments. Therefore, for the CA/xylitol cross-linking process with five independent parameters (n = 5), the required number of experiments is
N = 2n + 2n + nc = 25 + (2 × 5) + 8 = 50
(2)
The whole design consisting of the 50 randomized experiments and the corresponding responses is presented in Table S1 in the Supporting Information. Cost-effectiveness Evaluation. Cost-effectiveness was evaluated for CA/xylitol and low formaldehyde DMDHEU (F-ECO) resin as well as other four formaldehyde-free finish resins including SRD787, WFF, NF, and BTCA based on the corresponding WRA measurements and cost data obtained from the Alibaba Group. The effectiveness indices were calculated based on the performance of the corresponding finished fabrics as described in the Supporting Information (Table S2−S3). EHS Evaluation. For CA/xylitol and DMDHEU cross-linking system, EHS criteria were integrated by means of an environmental and human health risk assessment developed by Koller et al.28 Metrics representing risks for humans and environment resulting from the chemicals used in two cross-linking systems were included for the assessment. All the parameters for a given chemical were selected from the corresponding Material Safety Data Sheet (MSDS) or Screening Information Data Set (SIDS). The key parameters for safety evaluation were Flammability (F) and Explosive (E). The key parameters for health evaluation were Acute toxicity (AT), Skin sensitization (SS), Eye effect (EE), Mutagenicity (M), Carcinogenicity (C), Reproductive (R), and Target organ systemic toxicity (TO). The key parameters for environmental evaluation were Aquatic toxicity (T), Persistency (P), and Bioaccumulation (B). All chemicals were classified into five groups remarked by different colors.
■
RESULTS AND DISCUSSION Polynomials Predicted by RSM for the CA/Xylitol Cross-linking System. The three most important criteria in assessing the quality of cotton fabrics finished with cross-linking resins are WRA, WI, and TS. Each criterion was modeled as the response to five variable finishing parameters. A reduced cubic model was used for predicting the responses as functions of the variables. Table 3 summarizes the fitted polynomial equations for each measurable response in terms of coded variables. The corresponding results from the ANOVA are also presented in Table 3. The ANOVA results indicate that the models are significant at 95% confidence level for all three responses, because all p values of regression are less than 0.0001. In addition, the adequate precision (AP) values are well above the desirable value of 4.29,30 Both the regression coefficients (R2)
NWS = a WI/WI 0 + bTS/TS0 + c(WRA − WRA 0)/WRA 0 (1) where WI0, TS0, and WRA0 are the corresponding values of the untreated cotton fabric. The weighting factors a, b, and c were set to be C
DOI: 10.1021/acssuschemeng.5b01213 ACS Sustainable Chem. Eng. XXXX, XXX, XXX−XXX
Research Article
ACS Sustainable Chemistry & Engineering Table 3. ANOVA for the Three Measurable Responses Predicted by the Reduced Cubic Modela final equation in terms of coded factors
response WI WRA TS
F value
AP
SD
R2
Radj2
LOF
b
226.64
51.69
0.90
0.9931
0.9887
0.4343c
