Research Highlights

Electromigration is one of the most critical problems limiting the future scaling of integrated circuits (ICs).  As more transistors are packed into ever smaller volumes, the total length of interconnecting wire is increasing as the wire simultaneously becomes narrower and more vulnerable to failure. Surprisingly, while chip fabs are presently churning out microprocessors at the “5-nm process node” with correspondingly tiny interconnects, before recent work by a STROBE team no measurements of electromigration pressures had been reported at length scales smaller than 10 μm.

Using electron-beam lithography, the  team fabricates tiny aluminum nanowires on electron-transparent silicon nitride membranes. They then apply an electrical current density of 10⁸ A/cm²,  which is enough to drive electromigration, and map the nanowire’s density with parts-per-thousand precision using electron energy loss spectroscopy (EELS) in a scanning transmission electron microscope (STEM).  Two separable effects appear.  First, thermal expansion decreases the wire’s density as Joule heating raises its temperature. Using aluminum’s known coefficient of thermal expansion, one can convert the density map into a temperature map with the same spatial resolution.  Second, via the electron “wind force” the electrical current simultaneously compresses one end of the wire while it tensions the other. Using aluminum’s known bulk modulus,  one can convert the density map into a pressure map.  The temperature effect is independent of the direction of the current, while the pressure effect changes sign with the current’s direction, which allows the two effects to be separated.  Thus the thermodynamic state variables T and P can be mapped with high spatial resolution everywhere inside the wire. The ability to simultaneously see the prompt thermal and electron-wind-force effects in nanoscale interconnects opens entirely new avenues for studying electromigration, a key technological problem in the microprocessors powering modern computing devices.

Liquids and solids are two fundamental states of matter. Although the structure of crystalline solids has long been solved by crystallography, our understanding of the 3D atomic structure of liquids and amorphous materials remained speculative due to the lack of direct experimental determination. Now, a collaborative team from UCLA, Lawrence Berkeley National Lab and Brown University has advanced atomic electron tomography to determine for the first time the 3D atomic positions in monatomic amorphous materials, including a Ta thin film and two Pd nanoparticles. Despite different chemical composition and synthesis methods, they observed that pentagonal bipyramids are the most abundant atomic motifs in these amorphous materials. Contrary to traditional understanding, most pentagonal bipyramids do not assemble icosahedra, but are closely connected to form networks extending to medium-range scales. Molecular dynamics simulations further revealed that pentagonal bipyramid networks are prevalent in monatomic metallic liquids, which rapidly grow in size and form more icosahedra during the quench from the liquid to the glass state. These results expand our fundamental understanding of the atomic structure of amorphous solids and will encourage future studies on amorphous-crystalline phase and glass transitions in non-crystalline materials with three-dimensional atomic resolution.

A team of physicists and engineers at CU Boulder have solved the mystery behind a perplexing phenomenon in the nano realm: why some ultra-small heat sources cool down faster if you pack them closer together. The research began with an unexplained observation: a team led by Margaret Murnane and Henry Kapteyn at JILA were experimenting with metallic nanolines on a silicon substrate that when heated with a laser, something strange occurred. Nanoscale heat sources do not usually dissipate heat efficiently. But if you pack them close together, they cool down much more quickly.

Now, the researchers know why it happens. The team joined forces with a group of theorists led by Mahmoud Hussein in Aerospace Engineering Sciences to use computer-based simulations to track the passage of heat from their nano-sized bars. The simulations were so detailed that they could follow the behavior of each and every atom in the model—millions of them in all. They discovered that when they placed the heat sources close together, the vibrations of energy (called phonons) they produced bounced off each other more efficiently when other heat sources were nearby, scattering heat away and cooling the bars down.

The group’s results highlight a major challenge in designing the next generation of tiny devices, such as microprocessors or quantum computer chips. When you shrink down to very small scales, heat does not always behave the way you think it should.

2D lateral and vertical heterostructures have been actively studied for fundamental interest and practical applications. Although aberration-corrected electron microscopy and scanning probe microscopy have been used to characterize a wide range of 2D heterostructures, the 3D local atomic structure and crystal defects at the heterostructure interface have thus far defied any direct experimental determination. Now, a collaborative team from UCLA, Harvard University, MIT and UC Irvine demonstrates a correlative experimental and first principles method to determine the 3D atomic positions and crystal defects in a MoS₂-WSe₂ heterojunction with picometer prevision and capture the localized vibrational properties at the epitaxial interface. They observe various crystal defects, including vacancies, substitutional defects, bond distortion and atomic-scale ripples, and quantitatively characterize the 3D atomic displacements and full strain tensor across the heterointerface. The experimentally measured 3D atomic coordinates, representing a metastable state of the  heterojunction, are used as direct input to first principles calculations to reveal new phonon modes localized at the heterointerface, which are corroborated by spatially resolved electron energy-loss spectroscopy. In contrast, the phonon dispersion derived from the minimum energy state of the  heterojunction is absent of the local interface phonon modes, indicating the importance of using experimental 3D atomic coordinates as direct input to better predict the properties of heterointerfaces. Looking forward, it is expected that the ability to couple the 3D atomic structures and crystal defects with the properties of heterostructure interfaces will transform materials design and engineering across different disciplines.

Understanding and controlling the forces that drive the formation of symmetry-broken phases in quantum materials is a key challenge in condensed matter physics. The nature of the correlated interplay between charge, spin, and lattice degrees of freedom, however, often remains hidden in equilibrium studies where adiabatic tuning masks the causal ordering of rapid interactions. This motivates the use of ultrafast electron diffraction (UED) to capture structural dynamics on intrinsic time scales, for insight into the role of atomic-scale lattice distortions and vibrational excitations in driving, stabilizing and ultimately controlling emergent phases.

In a recent publication, a STROBE team carried out the first-ever ultrafast study of tantalum telluride (TaTe₂), yielding direct insight into its structural dynamics. The material exhibits unique periodic charge and lattice trimer order, which transitions from stripe-like chains into a (3×3) superstructure of trimer clusters at low temperatures. After cooling to 10 K, the thin crystalline films were optically excited and the structural dynamics was probed with the high-brightness electron bunches. Satellite peaks as well as sign changes in the complex diffraction patterns yield a fingerprint of the periodic order and structural transition. Our experiments captured the photo-induced melting of the trimer clusters in TaTe₂, evidencing an ultrafast phase transition into the stripe-like phase on a ~1.4 ps time scale. Subsequently, thermalization into a hot cluster superstructure occurred. Density-functional calculations indicate that the initial quench is triggered by intra-trimer Ta charge transfer, which destabilizes the clusters unlike CDW melting in other TaX₂ compounds.

Critical to this project were new methods and algorithms, enhanced microscopes and samples, advanced sample preparation as well as a unique high repetition rate ultrafast electron diffraction beamline utilized by the STROBE team from UC Berkeley, LBNL and UCLA

Understanding the chemical and physical properties of surfaces at the molecular level is highly relevant in the fields of medicine, semiconductors, batteries, etc. where precise atomic level control of determines materials and device performance. In particular, molecular order and domains affect many of the desired functional properties with carrier transport, wettability, and chemical reactivity often controlled by intermolecular coupling. However, both imaging of molecular surfaces and spectroscopy of molecular coupling has long been challenged by limited chemically specific contrast, spatial resolution, sensitivity, and precision. In this work, a team of  STROBE researchers demonstrate vibrational excitons as a molecular ruler of intermolecular coupling and quantum sensor for wave function delocalization to image nanodomain formation in self-assembled monolayers. In novel precision spatio-spectral infrared scattering scanning near-field optical microscopy combined with theoretical modelling few nanometer domain sizes and their distribution across micron scale fields of view could be resolved. This approach of vibrational exciton nanoimaging is generally applicable to study structural phases and domains in a wide range of molecular interfaces and the method can be used for engineering better molecular interfaces, with controlled properties for molecular electronic, photonic, or biomedical applications.

Scattering scanning near-field optical microscopy (s-SNOM) provides for spectroscopic imaging from molecular to quantum materials with few nanometer deep sub-diffraction limited spatial resolution. However, conventional acquisition methods are often too slow to fully capture a large field of view spatio-spectral dataset. Through this collaboration, STROBE researchers, at CU Boulder and the ALS –Berkeley, demonstrated how the data acquisition time and sampling rate can be significantly reduced while maintaining or even enhancing the physical or chemical image information content. The novel data acquisition and mathematical concepts implemented are based on advanced data compressed sampling, matrix completion, and adaptive random sampling. This research is of particular interest in synchrotron based nano-imaging facilities. This work paves the way to true spatio-spectral chemical and materials nano-spectroscopy with a reduction of sampling rate by up to 30 times.