For decades, understanding the origin of high temperature superconductivity has been regarded as the Holy Grail by physicists in the condensed matter community. The importance of high temperature superconductivity resides not only in its technological promises, but also in the dazzling number of exotic phases and elementary excitations it puts on display for physicists. These myriad phases and excitations give physicists new dimensions and building bricks for understanding and exploiting the world of collective phenomena. The pseudogap, charge-density-wave, nematic and spin liquid phases, for examples, are a few exotica that are found in cuprate high temperature superconductors. Understanding these phases is important for understanding the mechanism behind high temperature superconductivity, but they are also interesting in and of themselves.
The charge-density-wave (CDW) phase in the cuprates – a spontaneous emergence of a periodic modulation of charge density in real space – has particularly garnered a lot of attention. It emerges upon the destruction of the parent antiferromagnetic Mott insulating phase with doping and it appears to directly compete with superconductivity. Whether or not these features are generic, or maybe even necessary, for high temperature superconductivty is an important question. Unfortunately, currently there exists no other comparable high temperature superconducting materials family that enables such questions to be answered.
Recently, the iridates have emerged as a possible analog to the cuprates. The single layer variant Sr2IrO4, for example, exhibits signatures of both a pseudogap phase and a high temperature superconducting phase. However, with an increasing parallel being drawn between the iridates and the cuprates in terms of their electronic phases, CDW has so far eluded detection in any iridate, calling into question the validity of this comparison. Rather than studying the single layer variant, we decided to look at the bilayer iridate Sr3Ir2O7 in which a clear Mott insulator to metal transition has been reported with doping.
While CDW has been observed in many materials, what made it elusive in cuprates for many years is its spatially short-ranged (it extends only a few lattice spacings long) and often temporally short-ranged (it blinks in and out of existence quickly) nature. To get a good view of this order, experimentalists had to literally pin it down using external influences like magnetic fields or chemical dopants to suppress the temporal fluctuations and then use very sensitive diffraction or scanning tunneling based probes to observe them.
But rather than looking in real space for signatures of the CDW order, an alternative approach is to look for them in the time domain. Works by the Gedik group at MIT and the Orenstein group at U.C. Berkeley have shown that one can use ultrafast time-resolved optical reflectivity to “listen” for the tone of a CDW to infer its presence in the cuprates. In these experiments, one impulsively excites a coherent mode of the CDW using a femtosecond laser pulse, much like one would excite the vibrational mode of a tuning fork by impulsively banging it. One then stroboscopically looks for these CDW oscillations via temporally periodic modulations in its optical reflectivity, much like one would listen for the tone produced by the tuning fork. If you manage to hear the tone of the CDW, then you have established its existence!
We applied a similar approach to Sr3Ir2O7 and its doped versions [hear our experiment]. To our delight, the ringing of a CDW mode sounded immediately upon doping across its Mott insulator to metal transition, implying that the electronic liquid born from the doped Mott insulator is unstable to CDW formation, very similar to the case in cuprates. Also like the case of cuprates, this charge-density-wave is of a special nature: it is either very short-ranged, or temporally fluctuating. Whether or not there is a superconducting phase that competes with the CDW in Sr3Ir2O7 remains to be seen. If so, the phenomenology of the cuprates may really be quite generic. If not, the interesting question of why not is worth pursuing. And who knows, maybe the fact that we have a system that can be controllably tuned between the antiferromagnetic order and the CDW order may find use in technology some day.