Lilian C. Hsiao

Safa Jamali

2025 Metzner Awardee

Northeastern University

Northeastern University weblink
Citation

For contributions to the computational simulation and physical understanding of the rheology of dense suspensions and attractive colloidal systems, and for pioneering the development of machine learning techniques and physics-informed data-driven approaches to rheology and our understanding of complex multicomponent systems.

Prof. Safa Jamali (Northeastern University) has made a wide range of multi-faceted and diverse contributions to the science and practice of rheology: from theoretical and fundamental contributions in suspension rheology, bringing advances in AI/ML technologies to rheology in a meaningful way, and to developing computational and data-driven tools that are of utility to a broad spectrum of rheologists. Safa was born in the city of Sari in northern Iran, the youngest of three brothers. He followed his older brother Sina into the area of polymers, and pursued an undergraduate degree in Polymer Engineering at Amirbakir University. Following his graduation in 2008 he applied to the inter- European Erasmus educational program in rheology and moved to U.C. Louvain where he studied in his first year with Prof Christian Bailly. Safa recalls doing a lot of extensional rheology measurements together with Florian Stadler who was a postdoc there at the same time, as well as extensive constitutive modeling conversations with Evelyn van Ruymbeke who was also then a postdoc in Louvain. For his second Erasmus year, Safa moved to Univ Minho in Portugal where he joined Jose Covas’s polymer processing laboratory and completed a fully experimental masters’ thesis project working with extensional and capillary rheometers. Having earned his master’s degree in Engineering Rheology in 2010, Safa moved to the US to work with Joao Maia who had relocated from Minho to Case Western Reserve University just a year earlier. Safa was the very first graduate student to work with Joao in the general area of adapting Dissipative Particle Dynamics (DPD) techniques to simulate the nonlinear flow of rheologically-complex systems such as dense colloidal dispersions and he was awarded his Ph.D. in Macromolecular Science and Engineering in Spring 2015. His work at CWRU resulted in seven papers, including lead author contributions to articles in Soft Matter, J. Chem. Phys. and J. Rheology. DPD is an interesting tool for rheologists because it enables one to span much larger length scales and timescales than is possible with direct molecular dynamics techniques, but the challenge is to develop appropriate potential interaction functions between the rather nebulous ‘dissipative particles’ that can capture, in a quantitative way, the response of real materials. Safa worked on selecting appropriately soft and long-range potentials and this theoretical work was heavily guided by a (very well-cited) collaboration with Norm Wagner’s group at U. Delaware who contributed their extensive experience in experimental colloid science and quantitative rheometry of such systems.

After graduating in May 2015, Safa came to MIT to work jointly with Prof. Bob Armstrong in ChE and me on an industrial-sponsored project funded by Chevron Technology Corporation in the broad area of “Flow Assurance” in the oil & gas sector. Our sponsors were interested in understanding the development of yield stresses, unyielded plug-like regions and other complex soft solid structures in waxy crude oils as they experience a complex thermal and flow history during extraction and pumping from the reservoir to the collection/refining facility. These materials exhibit a set of characteristics that are now commonly called ‘elasto-viscoplastic’ (EVP) in character and they are also strongly thixotropic (i.e., they show a strong dependence on the entire flow history). They thus present a serious challenge to fluid mechanicians and rheologists interested in trying to simulate the flow of such materials and at that time were not readily described by existing closed-form constitutive theories – which typically focused exclusively on the ‘simpler’ cases of inelastic/ thixotropic materials or viscoelastic fluids such as polymer solutions and melts, or viscoplastic materials such as muds/clay dispersions (but not all three at once). EVP modeling has become a hot topic in the world of complex fluids in part because of the rapid and ongoing development of high-energy density flowable battery slurries as well as multiphase fracking fluids and drilling muds etc.

Safa was at MIT for approximately two years and made a number of important contributions that appeared in a series of three papers (two PRLs and a Materials Today Advances publication). He developed a modified DPD algorithm that allowed him to use soft/long-range Morse potentials to simulate dense clusters (of many 1000s) of hexagonal and weakly-attractive wax microcrystals, over a wide range of timescales, interaction strengths and imposed deformation conditions. His simulations were able to capture much of the complex rheophysics that these materials demonstrate experimentally; including fractal cluster formation and percolation (at very low volume fractions) to form a soft gel-like structure with a static yield stress; as well as elasto-viscoplastic (EVP) response characteristics in small and large deformations. His techniques enabled us to quantify the dynamical evolution in cluster size, shape and orientation under shear – which is manifested globally as the ubiquitous ‘thixotropy’ measured by experimentalists. The time-dependent simulations also show the onset and growth of spatial instabilities in local number density concentrations that nucleate, advect and grow to form the “shear-bands” that are of great interest to many complex fluid physicists studying materials as diverse as frictional granular flows, monodisperse polymer melts and even wormlike micellar systems. Subsequently Safa also adapted emergent ideas from the granular mechanics community such as computation of a “fabric tensor” which parameterizes and quantifies the particle-level connectivity of the complex sample-spanning ensemble structures that percolate and evolve in the system. His calculations showed that by following the evolution in magnitude and orientation of this fabric tensor we could connect microstructural measures to the simpler and empirical “structural thixotropy parameter” usually favored by experimentalists. While at MIT, Safa also interacted frequently with both Prof. Ken Kamrin in my department and also with the late, great Jim Swan in ChE, who were both interested in related topics in wet granular flows and in colloidal gels respectively.

He has continued to grow and develop these ideas after launching his own career in the Department of Mechanical Engineering at Northeastern University in 2017. In particular, Safa has pioneered a number of data science applications of Machine Learning to rheometry – through a large collection of papers including articles in J. Rheol., a highlighted “young researcher” special feature issue of Rheol. Acta as well as several articles published in Proc. Nat. Acad. Sci. in 2023 and 2024 with his experimental collaborator, and previous Metzner Award winner, Simon Rogers (ChE, UIUC). These latter articles introduced the idea of a ‘digital twin’ of a complex fluid in silico with parameters that could be “learned” (or best fit) by regression against a wide range of input rheometric data. His body of work as a junior faculty member at Northeastern University has brought AI/ML to complex fluids and rheology, and developed AI tools as robust methods for complex fluid modeling. This work has taken several pathways: from construction and detection of unbiased constitutive models for different complex fluids, to developing multi-fidelity platforms for highly accurate predictions of the rheological behavior of a given system through a general platform called “rheology-informed machine learning”. As a pioneer of datadriven techniques in rheology, Safa also wrote a perspective article in the 2023 Rheology Bulletin, co-edited a special issue of Rheologica Acta and also recently published a 2024 review in Current Opinions in Colloids and Interface Science. He continues to spearhead much of our community’s efforts in this direction.

Finally, I think it is worth noting how much service time Safa devotes to the rheological community. He has been a staunch supporter and organizer of the Rheology Research Symposium since its inception and also serves as Chair of SOR’s Membership Committee. He co-organized SOR’s extremely successful Covid-era online workshop on the physics of dense suspensions in July 2020 with Emanuela del Gado and Jeff Morris, and was also the Technical Program Co-Chair (with Kendra Erk) last year in Austin, TX. In his spare time (such as it is!) Safa enjoys music and the arts. He plays the hang drum as well as a number of other musical instruments. Like me, he describes himself as a Scotch enthusiast (with a slowly, but steadily growing single malt collection), a decent poker player (unlike me!), and a half-decent cook (fusing different flavors of eastern and western cuisine together). Like many of us he notes that one of the biggest perks of this job is the chance to travel to so many different places, and interact with students and collaborators around the world. His enthusiasm for our field and for SOR is contagious as I hope you will discover during his 2025 award lecture.