by Natalie Brown
Recently I performed Zero-span measurements on a selection of 19th and 20th Century Chinese papers. A Zwick Roell tensile strength machine, equipped with a modified z span application, was used on a sample set that was a mixture of traditional handmade paper (image A) and mass-produced machine made papers (image B). Before I began I had no experience performing z span measurements, what I thought would be a straightforward experiment turned into a week of preliminary testing and results that sometimes had me scratching my head.
A little introduction: Z span is an established method for measuring fibre strength, used widely in the paper making industry, and has been adopted by conservation researchers for degradation and characterisation studies since the 1980s. It is a form of tensile strength testing (TS), however the results reflect the properties of the individual fibres, rather than the sheet, due to the close proximity of the clamps (image C).
Z span is an attractive alternative to TS for low-density handmade papers, where fibres are randomly orientated. For me it was no different, many historic Chinese handmade papers were constructed of a loose network of long, intertwining fibres with little or no sizing (image D). The role that fibre-to-fibre bonding plays in these papers’ ultimate tensile strength is not fully understood and is still a topic up for debate (Mark and Habeger: 2001 pg. 329). In preliminary tearing tests, the handmade Chinese papers had very low inter-fibre bonding but high fibre strength. Measuring fibre strength was therefore thought to be a better reflection of the paper’s physical properties. The test also uses less material; I was able to successfully measure 70X15mm strips as apposed to 120X15mm stripes needed for TS. On the widely used Pulmac Zero Span models, it is my understanding that only small strip a few centimeters long is needed for up to 12 measurements.
Of course, no method is perfect and the methodological development for z span was taxing. Although there are standards for the paper making industry, within conservation research the method is not a standardized procedure. After consulting both TAPPI 231 cm-96 and ISO 15361:2000 standards, as well as studies that have used z span, there was still a big question unanswered for my own research; if the grain direction is not clear, where do I measure?
My biggest concern for the tests was repeatability and a high variance between measurements due to the unequal distribution of pulp in the paper. There are many options on where you can measure in a paper, but as historical samples were being used material was in short supply. Luckily, the University of Iowa Libraries research project “Paper Through Time” explored several options on where to take a measurement. They found measuring parallel to the chain lines yielded higher values in all but one paper, the researchers believe this is due to a grain direction created by the action of the mould going into the vat of pulp during sheet formation (Barratt; 2015). My own tests confirmed their findings, but the results also had a higher variance and standard deviation. I also observed that in many papers there was an increased concentration of fibres along the chain lines. The higher density of pulp along the chain lines was not representative of the paper, and could have be the reason for the higher variance. It was decided that measuring parallel to the laid lines, against the chain lines (in Iowa’s research, tab ‘W’); 5mm away from chain lines, to exclude excess pulp, would be the most representative and homogenous area of the paper. Although this decreased the values it also decreased the SD and variance in all papers where both directions were tested.
Of course for degradation studies it is more beneficial to have higher values, as higher values will produce a greater difference in before and after tests, and maybe the variance isn’t as important if you’re testing the same paper. However, for characterisation studies between different samples it is better to have values that are representative of the paper as a whole, and a higher variance would mean that the average result is not accurate. Like many experiments, where to measure for z span comes down to your need and in what capacity the results will be used. What has become clear, however, is that Z Span, within the context of measuring handmade papers, is not a standardized method, regardless if a standard is named in the paper. Iowa’s research has shown that where you choose to take a sample will affect the results of the test. To make experiments reproducible, and for people to avoid excessive and wasteful preliminary tests on historic samples, I think there needs to be more information relating to methodological development, like sampling location, included within research papers.
Barrett, Tim: email correspondence May 2015
Barrat, Tim et. al: Paper Through Time, Nondestructive Analysis of 14th-through 19th-Century Papers http://paper.lib.uiowa.edu/overview.php Dec. 2012
Mark, Richard; Habeger, Charles: Handbook of Physical Testing of Paper, Vol. 1 CRC Press 2001
TAPPI 231 cm-96 Zero-span breaking strength of pulp (dry zero-span tensile) 1996
ISO 15361:2000 Pulps. Determination of zero-span tensile strength, wet of dry 2001