Student Perspective: From Atoms to Algae: Revealing Nature’s Quantum Tricks with a Quantum Computer With Mingyu Kang

Are there quantum effects in biology? Together with various biological functions such as vision, olfaction, DNA mutation, and consciousness, photosynthesis is widely studied through the lens of quantum mechanics. Let’s take a look at a light-harvesting complex (LHC), an aggregate of pigment molecules found in photosynthetic organisms. As the LHC harvests energy from light, the energy excitation “hops” between molecular sites and is eventually delivered to the reaction center, where the energy releases electrons and protons that eventually produce sugar in plants. This energy transfer occurs in less than a picosecond with remarkably high efficiency. The molecules in the LHC are intricately connected, enabling energy to be transferred through multiple pathways. This leads us to ask, “Does Nature harness quantum effects, such as interference of pathways and entanglement between molecular degrees of freedom, to make photosynthesis more efficient?” 

To answer this question, the LHC is described by a quantum model that allows for interference and entanglement to occur. Then, simulating the dynamics of the model may reveal the relationship between the quantum effects and the energy transfer efficiency. However, simulating such a model using conventional computers requires prohibitively long runtime and large memory. One reason for these challenges is that a fully quantum model of the LHC includes not only the electronic states of the molecules but also their vibrations. The quantum vibronic (vibration + electronic) model is written as a very large matrix whose size grows exponentially with the number of electronic states and the vibrational modes. 

To overcome this computational difficulty, trapped ions, one of the leading platforms for building a quantum computer, may come into play. In a trapped-ion quantum computer, the stable atomic levels of each ion are used as a qubit. But out of the spotlight, there are also the motional modes, nanometer-scale oscillations of the ion chain due to quantum fluctuations. While these motional modes are typically used to mediate entanglement between the trapped-ion qubits, they can also be used as ingredients for the quantum simulation of molecular systems. The molecular system’s electronic states can be mapped to the trapped-ion qubits, and the vibrations can be mapped to the motional modes. This one-to-one mapping circumvents dealing with exponentially large matrices that made simulations on conventional computers intractable. 

Image

Duke University researchers in quantum computing and theoretical chemistry joined forces to see whether trapped ions can achieve quantum advantage; that is, whether trapped ions can simulate large molecular systems that cannot be simulated using conventional computers. This work, published as a review in Nature Reviews Chemistry [1], compared the expected accuracy and runtime of a state-of-the-art trapped-ion quantum simulator to those of leading simulation methods on conventional computers. 

“Where the review shines for me is that it does not stop at making hand-waving arguments on how the quantum simulation may outperform classical methods, but rather attempts to find out where it actually will,” said Hartmut Häffner, professor in physics at UC Berkeley (who was not involved in this work). “For instance, quantum advantage is defined. The definition may look like a trivial thing, but it is important!”

The review points out that an advantage of trapped ions is the all-to-all connectivity: each qubit can interact with all other qubits, and each qubit can interact with nearly all motional modes of the ion chain. This is a powerful feature for simulating molecular systems, whose complicated connectivity often makes simulations on conventional computers even more challenging. 

However, a challenge of trapped-ion quantum simulations is noise. Despite the remarkable experimental progress, stabilizing and precisely controlling the ion chain’s tiny quantum oscillations remains a difficult task. The review estimated that a trapped-ion simulator with a few tens of ions can achieve quantum advantage when noise is reduced by a factor of 10 to 100 from the state-of-the-art system. 

Instead of fighting against noise, Duke researchers turned the table and looked for ways to use noise in trapped ions as an ingredient for quantum simulations. They recognized that vibrations in molecular systems are also noisy. For example, spectroscopy of a molecular system reveals broadening of the peaks that represent the vibrational modes. The broadening turns out to be analogous to noise that occurs in the motional modes of trapped ions. This led to a novel experimental method of simulating molecular systems by injecting noise into the control of motional modes. As noise is used as an ingredient, this quantum simulation method is naturally robust to the trapped-ion system’s noise. 

The Duke team used a system of 7 trapped ions to simulate molecular energy transfer. The team developed experimental protocols for precisely controlling the motional modes. Combined with the method of injecting noise, the team achieved a versatile tunability of the simulated spectrum of the vibrational modes. As the method is robust to noise, the simulated energy transfer matched the theoretical predictions remarkably well. This experiment, led by RQS graduate student Ke Sun (now a postdoc at UC Berkeley), is described in a recent publication in Nature Communications [2].

“A quantum simulator as described in this work can explore quantum phenomena much better than they can be studied in real systems,” said Hartmut Häffner. “The reason is that one can modify the simulation and so systematically explore how the various microscopic properties influence the resulting phenomena, such as equilibrium population and energy transfer efficiency, etc. So in my view, it is not so much that it is easier to measure all sorts of quantities in a quantum simulator, but rather that one can change the system.”

The toy model of the molecular system simulated so far is still small and completely tractable using laptop computers. However, the noise-robust method demonstrated here is an important stepping stone towards simulating large molecular systems such as photosynthetic LHCs, beyond the reach of conventional computers. While noise is a challenge for quantum simulation, it is not necessarily a bottleneck, as quantum phenomena in Nature are also noisy. Characterizing noise processes in both the simulator hardware and the natural system could be a path towards accelerating practical quantum advantage.  Ongoing research in this area will enhance our understanding of how quantum interference and entanglement contribute to biological processes.

 

References:

[1] Mingyu Kang et al., Seeking a quantum advantage with trapped-ion quantum simulations of condensed-phase chemical dynamics, Nat. Rev. Chem. 8, 340-358 (2024)

[2] Ke Sun, Mingyu Kang et al., Quantum simulation of spin-boson models with structured bath, Nat. Commun. 16, 4042 (2025)