|[Home] [APPW 2004] [Journal papers]|
FLEXOGRAPHIC PRINTING ON TISSUE
S M Loffler, V J Dusting, N Vanderhoek and S Nikolovska
Presented at the 59th Appita Conference, Auckland, New Zealand 16-19 May 2006
Flexographic printing on a highly absorbent substrate such as tissue or towelling presents a unique set of technical challenges. The very absorbency that is critical to the enduse performance of the product is detrimental to print quality, yet demands for consistent, high quality printable results continue to grow.
This paper details work undertaken to understand factors affecting flexographic print quality on towelling substrates. Laboratory scale printing assessment has been used to characterise print performance in terms of colour density and print parameters. By looking at the interrelation between these parameters and ink physical characteristics such as viscosity, pH and surface tension, an understanding of the key properties of the inks required for this application has been obtained.
Equally important in the development of flexographic inks for printing on tissue and towelling is to ensure that product resistance factors such as ink bleed or dry rub removal are minimised. The development of reliable laboratory testing regimes that assess these properties in a way that reflect observations seen in consumer practice is therefore essential. This paper will report on test methods developed to assess the integrity of the ink-paper interaction and the relationship between the observations seen in laboratory analysis and in practical usage of the substrate.
Materials and Methods
Printing inks were supplied via Svenska Cellulosa Aktiebolaget (SCA) Tissue for analysis, from two major ink suppliers. In total, 35 inks were tested.
Four polymeric materials were provided by White and Gillespie as alternative flexographic printing plates. Trial plates for the IGT tester were made from each of these materials: A fifth plate material, sourced through the National Printing Laboratory was also used in some experiments.
The substrate used was an uncreped, single ply towelling sheet of basis weight 21 gsm.
Viscosity was measured using the number 2 Zahn cup at the National Printing Laboratory, which gives an effective viscosity in seconds by measuring the time for a known volume to flow through a small orifice at the base of the cup. Viscosity was also measured using a Brookfield viscometer with a spindle speed of 50 rpm, with the sample in a water jacketed container at a constant temperature of 25oC.
The pH of the inks was measured using the PTI-55 digital pH meter at Ensis Papro.
The surface tension of each ink was analysed using the Sensadyne Surface Tensiometer at the National Printing Laboratory. Surface tension was calculated using the maximum bubble method, where bubbles of gas are blown through two probes of different radii immersed in the ink, with the fluid surface tension being related to the differential pressure signal ΔP produced.
Dynamic Contact Angle
Dynamic contact angle was measured using the Fibro DAT 1100 at the National Print Laboratory. A drop volume of 0.5mL of the tested solution (water or ink) was dispensed by the DAT unit onto the surface of the substrate of interest. The machine then captures images of the drop in cross section every 20 milliseconds, and uses image analysis to calculate the drop height, surface area on the substrate, and contact angle.
Printing Trials – IGT tester
Ten printed ink strips on each substrate were produced using the IGT F1 tester at the National Printing Laboratory for each ink evaluated. Both Reflex and towelling substrate (as supplied by SCA) were tested.
The IGT tester was set up with the following configuration:
Anilox Force 100N
Printing Force 150N
Print Quality Analysis
Print colour density was measured using the X-Rite Model 408 Densitometer shown in Figure 1 located at the National Printing Laboratory.
Figure 1: X-Rite Model 408 Densitometer
Print quality was also measured using image analysis techniques. The printed substrate was scanned and saved as a TIFF format file at a resolution of 600 dpi.
Colour density was measured by taking a histogram of a solid colour area (such as the one illustrated in Figure 2), on a scale of 0 (colour) to 255 (white / no colour) for cyan, magenta , yellow or white, or from 0 (black / no colour) to 255 (colour) for red, green or blue. The colour density value was then calculated by finding the mean value of the histogram, and normalising the result on a 0-100 scale, where a value of 100 is a maximum intensity value for a given colour.
Figure 2 - Indicative solid colour block area
Colour variation was calculated by measuring the standard deviation in colour for the same histogram used to determine colour density.
Ink spread was calculated by measuring the number of coloured pixels in a nominal thick line region (dimensions 35 mm x 2 mm). The theoretical number of pixels at the 600 dpi resolution for this area is 39060; by measuring the actual number of coloured pixels, the tendency for this line to spread can be quantified.
The Sheen Wet Abrasion Scrub tester at the Australian Pulp and Paper Institute was used to test dye fastness on towelling and Reflex paper. The substrate (tested in duplicate) was taped to a sponge that fitted the arm of the tester. The arm was weighted with 400g, 1 cycle = 2 passes was set and approximately 1 mL of water was placed on the paper surface that would receive any ink that was transferred. Samples were ranked using the subjective scale given in Table 1.
Table 1 - Scuff Test rating scale
Further scuff test analysis was performed with commercially printed samples printed with two unidentified inks, and on products that had been identified as being of poor quality based on customer feedback. For comparison a hand-rub test was also performed by wetting the towelling pattern with water and applying two rubs on the surface of Reflex paper.
A plate cleanability test was developed to assess the difficulty of removing ink dried from a particular plate material. Ink (20 µL) was placed onto the surface of the plate, and allowed to dry overnight. The plate was then placed in a beaker containing water (50 mL) and allowed to stand for one hour, with occasional manual stirring. The plate was then removed from the water, and a UV-visible spectrum taken of the wash water with a Hewlett-Packard 8452A UV-Visible spectrophotometer. This spectrum was compared with the spectra of a solution of ink (20 µL) in water (50 mL) (representing complete removal of the ink) and the relative peak intensities of the two spectra used to determine a plate cleanability rating.
RESULTS AND DISCUSSION
In order to relate end print performance with ink quality it is necessary to characterise the inks in terms of fundamental parameters that are relatively simple to measure and also discriminate between the samples. A series of tests were undertaken to this end and are reported in turn below.
The Zahn cup, although commonly used by printers as a method of determining viscosity, provides little discrimination between samples. Of the samples measured with the Zahn cup, only two of the 35 inks had Zahn viscosities that were not in the range of 14-15 seconds.
The viscosities measured with the Brookfield viscometer had a much broader spread, with viscosities ranging from 2 to 49 cP. This would suggest that although the Zahn cup may be useful in a commercial sense for screening out the most problematic inks, the Brookfield viscometer is much more sensitive and better able to determine subtle variations in viscosity that may impact print performance.
Viscosities with the Brookfield viscometer were measured at a range of spindle speeds from 20 to 100 rpm. For most samples, the viscosity varied little with the spindle speed, with an increasing speed giving a slightly lower viscosity in most cases, as would be expected.
The pH for each of the inks was measured, with all inks falling in the pH range 7.8 to 12.2. Perhaps significantly, a number of the inks had pH levels above 9, which has been suggested in the literature  to lead to the printed ink not achieving adequate water resistance after it dries. Such printed inks might well bleed in water or even wash off the substrate several days or even weeks after printing.
Measured surface tension values fell in the range 29-42 mN/m. Repeatability was tested by performing 10 repeat tests for 3 of the inks, and standard deviations were between 0.1 and 0.9 for these samples.
Dynamic Contact Angle
Figure 3 shows the dynamic contact angle for the inks and the towelling substrate. It can be seen that most of the inks rapidly penetrate into the sheet, with contact angles going to zero (indicating disappearance of the drop from the surface of the sheet) in less than 0.5 seconds. There are two notable exceptions, both of which are still on the surface when more than 1 second has elapsed since the deposition of the ink on the sheet surface.
Figure 3 - Dynamic contact angle for measured inks on towelling substrate. Note that all but two inks penetrate into the sheet fully within one second.
Print Quality – Densitometry
Densitometry results for the inks were obtained from print tests using the IGT F1 tester on both the towelling substrate and Reflex, a printing paper used for comparison purposes. A higher value for ink density is indicative of more ink at the surface of the sheet and is a positive observation. The results from the towelling substrate are summarised in Table 2 below. It can be seen that ink B gives the highest values for colour density of the inks evaluated.
Table 2 - Colour density readings for Towelling Substrate printed on IGT F1 tester
Print Quality – Image Analysis
Figure 4 shows a graph of the colour density, colour variation and ink spread for the inks tested. The colour variation is indicated by the size of the bubble, with a large bubble indicating a large degree of colour variation (worse result). The ideal ink would have a high colour density, an ink spread value of 100, and a small colour variation. As can be seen from Figure 2, Ink "B" comes closest to meeting these criteria. The variation between the different colours for a given ink is also noteworthy.
Figure 4 - Colour density, ink spread and colour variation (indicated by bubble size) for evaluated inks.
Figure 5 shows the data from Figure 4 rearranged by the ink colour. The colour strength obtainable from the yellow inks printed on the towelling substrate is clearly the greatest in most tests, whereas the magenta in general gave the lowest colour density values.
Figure 5 - Colour Density, Ink Spread and Colour Variation arranged by ink colour.
Relationship between Colour Density and Ink Physical Properties
The relationship between the colour density obtainable from printing with the IGT tester and various physical print properties is illustrated in Figures 6 to 8. The colour coding in these graphs relates to the process colours of the inks evaluated. The "DAT-X-intercept" (Figure 8) is extrapolated from the dynamic contact angle data presented in Figure 3, and represents the time elapsed from when a drop of ink hits the surface of the sheet until it disappears from the surface.
Figure 6 - Relationship between printed colour density and ink surface tension.
Figure 7 - Relationship between printed colour density and ink viscosity.
Figure 8 - Relationship between printed colour density and DAT x-intercept.
It can be seen from Figure 6 that there is little correlation between surface tension and resultant print quality, though it is noteworthy that there is not a great range in surface tension data, with the majority of inks having surface tension values between 30 and 40 mN/m. Similarly, while the viscosities of most inks are less than 20 cP, the two outlying points in Figure 4, both of which are ink "B", relate to the highest colour density prints. The data in Figure 8 is not conclusive (related in part to the relatively quick absorption of all the inks into the substrates), but would suggest that an ink that remains on the surface of the sheet for a longer time would give a print with a higher colour density. Inks with higher viscosity best meet this criterion.
In order to further examine the relationship between print quality and viscosity, the highest viscosity ink was diluted with water to reduce the viscosity, and printed onto the towelling substrate with the IGT tester. The results of this work, given in Table 3, suggest that reducing the ink viscosity does have some impact on colour density. It must be noted, however, that diluting the ink with water will decrease the solids content proportionally, and that a better investigation of ink viscosity effects would use viscosity modifying additives without reducing solids content.
Table 3 - Relationship between viscosity and colour density for highest density ink
The results from the scuff tests are given in Tables 4 to 8.
Although reasonable variation was seen when the inks that had been printed in the laboratory on the towelling and Reflex substrates were tested (Table 4), the variation was not evident when samples that had been generated in machine printing trials were analysed (Table 5). Further, there was significant variation with samples that were reported by SCA to have scuff problems (Table 6) or had received customer complaints about poor scuff resistance (Table 7), with the seashells sample demonstrating no scuff in laboratory testing.
The results of the scuff test are therefore inconclusive, and not consistent with customer and anecdotal evidence of poorly performing products. Further analysis is necessary to develop a laboratory scuff analysis that discriminates good and poorly performing products.
Table 4 - Scuff Test Results – Laboratory Samples
Table 5 - Scuff Test Results – Machine Generated Samples
Table 6 - Scuff Test Results – "Poorly Performing" Samples (SCA anecdotal)
Table 7 - Scuff Test Results – "Poorly Performing" Samples (Customer Complaint)
IGT Printing Parameters – Printing Force
In order to evaluate the effect of the IGT printing force on print quality, two inks were selected for printing with the print force varying from 75 to 250N (compared to the default value of 150N). The results are given in Tables 8 and 9. It can be seen from these results that the colour density is greatest at the default print setting of 150 N, and that colour variation and ink spread do not vary greatly with the printing pressure.
The ability to select an optimum printing pressure is not surprising. Too little pressure would prevent adequate transfer of ink from the printing plate to the substrate, and too much pressure would be expected to force ink into the substrate, rather than keeping the ink on the surface where the apparent colour would be enhanced.
Table 8 - Effect of printing force on Reflex Print Quality
Table 9 - Effect of printing force on Towelling Print Quality
Plate Analysis – Print Quality
Print testing was performed on the IGT tester using two of the inks – "B" and "C" – using five plate materials (four supplied by White and Gillespie, and one sourced through the National Printing Laboratory). The colour density measurement for each of these trials is shown in Figure 9.
The colour density obtained when the WG1 plate is used exceeds the colour density obtained for the other plates by between 10-20%. This is a consistent observation over the range of process print colours and over both inks. The colour density for the "B" ink for a given plate material was also greater than for the "C" ink for the same material, consistent with earlier observations.
Figure 9 - Print colour density for tested plate materials for "B" and "C" Inks.
Plate Analysis – Contact Angle
The contact angle for each of the plate materials, measured with the "B" and "C" inks and with water as the investigating fluid is reported in Table 10. There is no clear trend in the relationship between contact angle and the print colour density as shown in Figure 9.
Table 10 - Contact angle (degrees) for plate and liquid Combinations
Plate Analysis – Plate Cleanability
Plate cleanability for a range of inks and plate materials was assessed using a methodology developed for this project (see "Materials and Methods" section). The results of the plate cleanability analysis were compared to anecdotal reports from SCA on the difficulty or otherwise of removing the ink from the plate material.
Table 11 details the measured plate cleanability values for the WG1 material for a range of inks, and compares these values with SCA reports on plate cleanability for a comparable material. It can be seen that the "B" ink, which was reported to be difficult to remove from the plates on the commercial Cobden-Chadwick printer at SCA, also proved difficult to remove from the plate in the laboratory. The "A" and "C" inks were easier to remove in commercial trials and both proved easier than "B" to remove from the plate in the laboratory. The "D" and "E" inks were significantly easier to remove from the plate than either of the Hydrofibe formulations in the laboratory test.
Table 11 - Plate cleanability assessment for "WG1" plate material
The cleanability of the alternative plate materials was also assessed, as reported in Table 12. There was no difference between the cleanability for the plate materials.
Table 12 - Ink removal assessment for alternative plate Materials
A number of factors affecting the print quality for flexographic printing of toweling substrates have been investigated in a laboratory study. The correct choice of ink can have a significant impact on print quality obtained, but may also impact factors such as runnability. Press conditions can be optimised by the selection of the printing plate material and press conditions such as printing pressure.
This work was carried out as part of the research program for CRC Smartprint. The authors wish to express their appreciation to the staff of SCA, and in particular to John Proctor, Technical Advisor, for their assistance and general advice during the duration of the project. The National Printing Laboratory and APPI also provided equipment essential to the completion of this work.
1. Todd, Ronald E. Printing inks - Formulations Principles, Manufacture and Quality Control Testing Procedures Pira Press, Leatherhead p. 299 (1994).
S. M. LOFFLER1, V. J. DUSTING2, N. VANDERHOEK3 AND S. NIKOLOVSKA4
1 Senior Research Scientist, 2 Technical Officer 3 Senior Principle Research Scientist, Ensis Papro, Private Bag 10, Clayton South MDC, Victoria 3169, Australia.
4 Development Leader, Technical Projects, Svenska Cellulosa Aktiebolaget, PO Box 117, Box Hill, Victoria 3128, Australia.