Characterizing the transplanar and in-plane water transport of textiles with gravimetric and image analysis technique: Spontaneous Uptake Water Transport Tester

Water absorption and transport property of textiles is important since it affects wear comfort, efficiency of treatment and functionality of product. This paper introduces an accurate and reliable measurement tester, which is based on gravimetric and image analysis technique, for characterising the transplanar and in-plane wicking property of fabrics. The uniqueness of this instrument is that it is able to directly measure the water absorption amount in real-time, monitor the direction of water transport and estimate the amount of water left on skin when sweating. Throughout the experiment, water supply is continuous which simulates profuse sweating. Testing automation could even minimise variation caused by subjective manipulation, thus enhancing testing accuracy. This instrument is versatile in terms of the fabrics could be tested. A series of shirting fabrics made by different fabric structure and yarn were investigated and the results show that the proposed method has high sensitivity in differentiating fabrics with varying geometrical differences. Fabrics with known hydrophobicity were additionally tested to examine the sensitivity of the instrument. This instrument also demonstrates the flexibility to test on high performance moisture management fabrics and these fabrics were found to have excellent transplanar and in-plane wicking properties.


Introduction
Supplementary

Components selection and specification
In order to minimise the amount of liquid that might transport within the fabric-plate interface, a non-wettable surface is preferred and a polytetrafluorethylene (PTFE) plate was selected. To enable automatic movement of the T-plate, a robust motor (980D28121, MFA) was used and its movement speed varies with the supply voltage.
The T-plate was raised at the speed of 2.7 mm/min (with the use of -3.5 V) while it was lowered at 6.9 mm/min (with the application of + 7.5 V). To ensure steady water level of the upper water tank, a micropump (D250-03, RS) with stable and relatively low flow rate (40 ml/min at maximum voltage supply) was chosen. To facilitate realtime monitoring of the water absorption amount in fabric, a balance (MS1003S, Mettler Toledo), with 0.001 g repeatability and maximum response rate of 22.9 data per second, was utilised. It is capable of connecting to the PC with RS-232C. To connect various electronic components with the computer for testing automation, data acquisition device (DAQ 6525, National Instruments) was used. It features digital input function for sensing DC voltage and digital output function for controlling the external voltage applied to motor for moving the T-plate. Two photoelectric sensors (EE-SX 671, ORMON) were wired to the digital input channels of the DAQ device. When the opaque plate reached the slot of the photoelectric sensor, a signal was generated and transferred to the computer via DAQ device. The feedback signal was then returned to DAQ device to disconnect voltage supply to motor. Therefore, the Teflon tube mounted on T-plate, with the aid of photoelectric sensor, was brought to the same height level for supplying water to the fabric in a repeatable manner.

Control programming
LabVIEW 2009 programming was used to develop an interface for controlling the movement of the T-plate, receiving input signal, performing logical decision, recording water uptake rate and analysing data. Figure S2 shows the control interface for SUWTT and it is divided into four parts as denoted by the number indicated in the figure. Part 1 is for inputting sample details, setting the testing duration, and controlling the start and end of the test. Part 2 is for recording the weight of the fabric before and after the test, and the mass of water absorbed by each layer can be calculated automatically. Part 3 is to capture the wetted pattern and by calculating the number of pixels in the recorded image, the wetted area was then known. Part 4 shows the time-dependence absorption curve of the sample in real-time.

Experimental details -Sample
The porosity of the fabric is calculated according to equation (1) with reference to Hsieh's work 10 . Supplementary

Operational Testing Procedures
Prior to the test, the instrument was 'warmed up' for 30 minutes and the ordinary testing procedure is listed below: -Set the water supply duration in the LabVIEW control interface (Part 1 of Supplementary Figure S2).

-
Check the water level in the bottom water reservoir and pour water to Section A of the upper water tank, if necessary.

-
Place the first specimen onto the sample podium and record its dry weight in the LabVIEW control interface (shown in Part 2 of Supplementary Figure S2) and repeat the same procedure for the remaining two layers if 3-layer test is to be conducted.

-
Apply an external pressure of 360 g (2.5 g/cm 2 ) onto the sample and tare the balance. -Press the 'start and record' button in SUWTT control interface (shown in Part 1 of Supplementary Figure S2). ▫ The program starts logging data. ▫ The T-plate starts moving up until the water from the orifice of Teflon tube approaches the sample.

Results and Discussion
Supplementary

Precision of the equipment
As the purposed instrument is based on gravimetric technique, the precision of the electronic balance (Mettler Toledo MS1003S), a dominant component for characterising fabric's absorbency in this study, should be reported and is found to be + 0.001g. Also, image analysis technique is adopted for characterising the wetted area. To ensure accurate measurement, the camera was mounted to a fixed position attached with constant lighting.

Uncertainty of different parameters
Some parameters such as water content and transplanar ratio cannot be measured in a single measurement. In fact, several physical factors are involved and the uncertainty of these parameters can be calculated from the uncertainty of each direct measurement 11 . The calculation of uncertainty of water content is based on the assumption that all errors are independent and random, and it is computed by the quadratic sum according to equation (2). For the transplanar ratio, the water absorption amount by top filter paper is divided by and interrelates with the absorption amount in the bottom filter paper, so the uncertainty of transplanar ratio is calculated by the sum of fractional uncertainties in two direct measurements, as shown in equation (3).
where δ is the measured uncertainty from each component.

Reproducibility of the SUWTT
In order to ensure the result is reproducible and reliable, it is found necessary to have a routine calibration of the entire set up. This may be done by testing a standard material for a pre-determined time and measuring for its weight and spreading area. A conventional filter paper with guarantee wetting performance was tested and the coefficient of variation (CV) of the collected amount and the spreading area for this standardised material is within 4 % for five replicates, it can be concluded that the setting is well calibrated.
To investigate the reproducibility of this test, the CV % of various measurement parameters is examined. For a 1-layer test, the average CV % of water absorption rate and wetted area for 15 types of fabrics is 6.086 and 3.632, respectively. For a 3layer test, the average CV % of amount of water absorbed by the bottom filter paper for 21 types of fabrics is 4.786. The results indicate that the reproducibility of the test is quite high. Comparatively, the CV % of the samples investigated is higher than that for the standardised material. It can be inferred that the fabric itself might contribute to the variance of the result since these samples are produced and finished in laboratory scale equipment, but not a commercialised set up. The CV % of AATCC Test method 79 -Absorbency of textiles also demonstrates the variability of these samples, suggesting the error of the result might come from the fabric itself.

Accuracy of SUWTT
In order to check the accuracy of the proposed set up, the SUWTT result was correlated with the one measured by the conventional measurement methods, such as wettability test (AATCC 79) 2 , vertical wicking test (AATCC 197) 3 , horizontal wicking test 12 , moisture management tester (AATCC 195) 6 and water absorption capacity test 12 . For easy and better understanding about these methods, the testing principles are summarised in Supplementary Table S7.
Supplementary A preconditioned strip of the specimen was suspended vertically with its lower end immersed in a reservoir of distilled water and the height of water reached in the fabric against gravity was visually observed and recorded after a fixed time. The initial and extended wicking rate, expressed in mm/s, indicates the average speed of water to reach 20 and 150 mm height, respectively.
-Initial wicking rate, 20 mm divided by time spent (mm/s) -Extended wicking rate, 150 mm divided by time spent (mm/s) Horizontal wicking test (Tang et al., 2014) A fixed amount of water was supplied at the bottom side of fabric at a constant rate (10 ml/h). A camera, standing on top of the set-up, was utilized to capture the image of the wetted sample and the water spreading area was measured.
-Horizontal wicking area (cm 2 ) Moisture management tester (AATCC 195) The sample was put in-between two sets of metal electrodes and a fixed quantity of liquid was dropped onto the back side of the fabric and the direction of water spread was traced automatically by the metal electrodes.
-Overall (liquid) moisture management capability (OMMC) Water absorption capacity test (Tang et al., 2014) Fabric was put into a tank of water and 5 minutes was allowed for it to sink completely into the water. The fabric was then taken out by tweezers and hung onto a rod vertically until there was no water dripping within a 30-second interval. The water gain in fabric was measured and it is expressed as mass of water gain per unit gram of fabric in percentage.
-Wet pick-up (%) The correlation results of various testing methods against SUWTT are illustrated in Figure S3 to Figure S6. Figure S3 shows that water absorption rate by SUWTT has moderate correlation with water absorption time by wettability test (Adj. R 2 =0.68), initial wicking rate by vertical wicking test (Adj. R 2 =0.69) and extended wicking rate by vertical wicking test (Adj. R 2 =0.69). Higher absorption rate under spontaneous uptake (i.e. by SUWTT) is associated with shorter water absorption time and higher vertical wicking rate. This direction of correlation is rational and the strength of correlation is acceptable, thus it suggests that water absorption rate is measured accurately. For the 1-layer SUWTT test, the absorption amount of the sample was recorded after 30 second water absorption. It is somehow related to the absorption capacity of the fabric and its correlation with the one by Tang et al.'s method 12 was studied. Figure  S4 illustrates that it has moderate and positive correlation with wet pick-up (Adj. R 2 =0.59) and this finding is understandable, implying the accuracy of this measurement.
Supplementary Figure S4  For the 1-layer SUWTT test, the wetted area of the sample was measured under demand wetting principle (i.e. the amount of water supply varies). This factor is affected by the wicking property and absorption capacity of the fabric. In order to standardise the measurement, the wetted area was divided by absorption amount of the sample and this was correlated with horizontal wicking area by Tang et al.'s 12 method. Figure S5 illustrates the correlation of wetted area by the two methods and these two factors are moderately correlated (Adj. R 2 =0.67).

Supplementary Figure S5. Wetted area per unit gram of water by SUWTT against horizontal wicking area
For the 3-layer SUWTT test, the transplanar ratio which reflects the transplanar wicking property of fabric was measured. Figure S6 demonstrates that it is strongly related to OMMC by MMT (Adj. R 2 =0.80). OMMC is an indirect measurement of moisture management property of fabric which is done by electrical method. A strong correlation implies that transplanar ratio by SUWTT gives accurate result.

Supplementary Figure S6. Transplanar ratio by SUWTT against OMMC by MMT
On the other hand, for the conventional testing methods, 'the water content of fabric' and 'the amount of water absorbed by the bottom filter paper' cannot be measured and so the accuracy of these two measurements cannot be verified. In brief, SUWTT can perform accurate measurement with comprehensive information provided under short testing time.