Eur. Phys. J. E 16, 77–80 (2005) DOI: 10.1140/epje/e2005-00009-x T HE EUROPEAN P HYSICAL JOURNAL E Non-equilibrium behavior of sticky colloidal particles: beads, clusters and gels H. Sedgwick 1 , K. Kroy 1, 2 , A. Salonen 1 , M.B. Robertson 1 , S.U. Egelhaaf 1 , and W.C.K. Poon 1, a 1 School of Physics, The University of Edinburgh, Mayfield Road, Edinburgh EH9 3JZ, Scotland, UK 2 Hahn-Meitner-Institut Berlin, Glienicker Str. 100, 14109 Berlin, Germany Received 17 November 2004 Published online 31 January 2005 – c  EDP Sciences, Societ`a Italiana di Fisica, Springer-Verlag 2005 Abstract. To understand the non-equilibrium behavior of colloidal particles with short-range attraction, we studied salt-induced aggregation of lysozyme. Optical microscopy revealed four regimes: bicontinuous texture, ‘beads’, large aggregates, and transient gelation. The interaction of a metastable liquid-liquid binodal and an ergodic to non-ergodic transition boundary inside the equilibrium crystallization region can explain our findings. PACS. 82.70.Dd Colloids – 87.15.Nn Properties of solutions; aggregation and crystallization of macromolecules Short-range interparticle attractions are ubiquitous in synthetic colloids. The equilibrium phase behavior of par- ticles with an isotropic short-range attraction is well known: the attractive interaction widens the fluid-crystal coexistence gap. The non-equilibrium behavior of sticky particles is far less understood. In the case of colloids, two qualitatively different glassy states exist at volume fractions φ  0.6 [1]. When φ  0.2 and the attraction is deep enough, permanent gelation is almost invariably observed. At shallower attractions (10k B T ), ‘transient’ gelation and separation into particle-rich and particle- sparse amorphous phases have also been described [2–4], together with report of a ‘cluster phase’ [5]. Theoretically, the gelation of sticky particles has long been discussed in terms of diffusion-limited cluster aggregation (DLCA) and percolation; recently mode-coupling theory (MCT) has been brought to bear on the problem [6]. In addi- tion, a metastable liquid-liquid phase boundary (binodal) exists. In polymer solutions, the interference of gelation and liquid-liquid phase separation leads to a rich ‘zoo’ of non-equilibrium behavior [7,8]. Here we investigate the non-equilibrium behavior of sticky colloidal particles, especially their aggregation and gelation, using a globular protein. Globular proteins in salt solution are often modelled as sticky particles [9–11]. This simple picture has already proven significant in equilibrium phase behavior: When the interparticle poten- tial is quantified by the second virial coefficient, a quasi- universal fluid-crystal coexistence boundary exists for syn- thetic colloids and proteins [12–14]. a e-mail: w.poon@ed.ac.uk As ‘particles’ we used a model globular protein, lysozyme. Adding sodium chloride reduces the inter- molecular electrostatic repulsion and hence the effective short-range attraction starts dominating [9–11]. Careful visual observations and microscopy delineate four non- equilibrium regimes. We give a unified description of these observations within a framework that should be applicable to other sticky-particle systems. We show that the precise kind of phenomena observable in a particular system is sensitive to a range of physical parameters. This helps ex- plain why a plethora of reported observations exists on apparently similar systems. Six-times crystallized chicken egg-white lysozyme (Seikagaku America) was directly dissolved in a 50mM sodium acetate buffer titrated with hydrochloric acid to pH = 4.5. Stock solutions at 150 mg/ml were centrifuged to remove dust and undissolved material. Higher concen- trations were reached using a Vivaspin 6 ml concentra- tor. Exact concentrations were determined by UV absorp- tion spectroscopy after 1000× dilution in buffer, using a specific absorption coefficient of 2.64 ml/mg.cm. (We cal- culate an effective φ using a molecular radius of 1.7 nm and molecular weight of 14320 g/mol.) Salt solution fol- lowed by deionized water were pipetted to the protein in buffer, mixed by shaking and observed by the naked eye at 22 ◦ C±2 ◦ C. Small amounts were drawn into 0.1mm-thick rectangular capillaries and observed using a Zeiss optical microscope. At increasing salt concentration, c s , samples remained homogeneous (◦, Fig. 1) until crystals (×) nucleated be- yond a crystallization boundary (CB). At higher c s , across a ‘non-equilibrium boundary’ (NEB), all samples turned