Secrets of a silky cotton shirt

Neutrons throw light on an important manufacturing process in the textile industry. 

You may have noticed that some of the highest quality cotton shirts or blouses have a soft silky sheen while also being hardwearing. This is because the cotton has undergone a process known as mercerisation. Cotton is more than 90 per cent cellulose, the world’s most common natural polymer, being an essential structural component of plants, as well as produced by some bacteria. In the mid-19th century, the British chemist John Mercer discovered that treating cotton fibres with a sodium hydroxide solution renders the fabric stronger and easier to dye.

This mercerisation process is still widely used today to enhance cotton textiles and other cellulose-based materials.

Studies indicate that mercerisation is a result of specific structural changes in the constituent cellulose fibres. However, what exactly happens at the molecular level has not been fully established – that is, until now. Using neutron fibre diffraction, we think we have uncovered the secrets of mercerisation.

Cellulose consists of microfibrils composed of long chains of thousands of glucose molecules, each linked through an oxygen unit. Glucose is a six-atom ring of five carbon atoms and one oxygen, with hydrogen atoms or hydroxyl (OH) units attached to the carbons. Because oxygen is strongly electron-attracting, this results in an uneven distribution of electric charge (polarity) across the rings, which is transmitted down the chains, resulting in an overall polar direction. Plant and bacterial cells generate bunches of these chains all with the same polar direction. They are held together by weak hydrogen bonds between specific interchain hydrogen and hydroxyl groups in a staggered way, so that they become tightly packed in parallel sheets that stack up to form a crystalline fibril. The fibrils further assemble into hierarchical structures such that those with opposite polarities may lie side-by-side, creating an overall crystal structure is known as cellulose I.


Chains on the move

When mercerisation is carried out, the polymer chains undergo a transformation that results in chains with opposite polar directions lying side by side within the same fibril so that its polarity is cancelled out. This crystal form is known as cellulose II. The change in crystal structure is induced by the sodium-hydroxide treatment which sucks water into the fibrillar structure causing it to swell and allowing the chains to loosen and move. Two explanations have been put forward as to how the chains switch to the antiparallel layout. In the first one, suggested by Francis Kolpak and John Blackwell (Case Western Reserve University, US), the interchain hydrogen bonds break allowing the constituent chains to migrate out of their fibril. They then aggregate with chains from a fibril of opposite polarity to form the antiparallel arrangement of cellulose II. The second theory, formulated by Henri Chanzy and E. J. Roche (CNRS, Grenoble) is more interesting. It suggests that the individual chains become free enough to fold back on themselves in a zigzag pattern to form crystalline regions of antiparallel chains within the same fibril.

Establishing which theory best explains this widely-used industrial process has been a longstanding challenge. Fortunately, neutron diffraction provides valuable insights. As in X-ray diffraction, neutrons are reflected from a crystal structure to give a characteristic diffraction pattern that allows structural characteristics to be determined. However, whereas X-rays do not ‘see’ hydrogen atoms very well, neutrons interact strongly to give a clear pattern. Furthermore, if some of the hydrogens are replaced with the heavier isotope of deuterium, which scatters differently from hydrogen and gives a different density pattern, it becomes possible to distinguish structural changes involving, for example, hydrogen bonding more clearly.


Using fibrils from a bacterium, we prepared samples of tiny crystals of cellulose I containing just hydrogen, and also of similar fibrils that were fully deuterated. We then made a random 50-50 mixture of the two, collecting the diffraction patterns of all three samples for comparison. The samples then underwent mercerisation to convert them into cellulose II, and the diffraction patterns again collected. By measuring diffraction intensities of the 50-50 cellulose II sample, and then comparing the results with what would be expected from the two theories in terms of the final distribution of the hydrogenated and deuterated chains, we concluded that the Chanzy–Roche model in which chains folded back on themselves in the same fibril offered the most likely description of mercerisation.  

ILL Instrument : Thermal neutron diffractometer for single-crystal and fibre diffraction D19


1. Sawada, D., Nishiyama, Y., Shah, R. et al. , Nat Commun13, 6189 (2022).

2. A. Buleon, H. Chanzy and E. Roche., J. Polym. Sci. Part C. Polym. Lett., 15, 265–270 (1977).

3. F. J. Kolpak and J. Blackwell, J.Polymer,19, 132–135 (1978).    


Research team: Paul Langan, Estelle Mossou and Masahisa Wada (ILL), Yoshiharu Nishiyama (ILL/CNRS), V. Trevor Forsyth (ILL/Lund University, Sweden/Kyoto University, Japan), Hugh Michael O’Neil, Daisuke Sawada and Riddhi Shah (ORNL, US)