The DRAG (Derivative Removal by Adiabatic Gate) technique is a sophisticated method employed in the control of transmon qubits, which are a type of superconducting qubit used extensively in quantum computing. The primary objective of this technique is to mitigate the effects of the Stark shift and to avoid unwanted transitions, which are critical for maintaining the fidelity and coherence of quantum operations.
Understanding the Stark Shift
The Stark shift, also known as the AC Stark effect, is a phenomenon where the energy levels of an atom or a qubit are shifted due to the presence of an external oscillating electric field. In the context of transmon qubits, this shift occurs when the qubit is driven by microwave pulses, which are used to manipulate its quantum state. The Stark shift can lead to detuning, where the qubit's energy levels are not aligned with the intended frequencies, causing errors in quantum gate operations.
Unwanted Transitions in Transmon Qubits
Transmon qubits are designed to have a weak anharmonicity, meaning that the energy difference between successive energy levels is not uniform. This weak anharmonicity makes transmons less sensitive to charge noise but introduces the risk of unwanted transitions to higher energy levels (e.g., from the ground state |0⟩ to the second excited state |2⟩, bypassing the first excited state |1⟩). These unwanted transitions can degrade the performance of quantum gates by introducing errors and reducing gate fidelity.
The DRAG Technique
The DRAG technique addresses these challenges by shaping the control pulses in a way that minimizes the effects of the Stark shift and suppresses unwanted transitions. The core idea of DRAG is to use a combination of amplitude and phase modulation in the control pulses. Specifically, DRAG involves adding a derivative component to the pulse shape, which compensates for the distortions caused by the anharmonicity of the transmon qubit.
Mathematical Formulation
Consider a Gaussian-shaped pulse used to drive a qubit transition. The DRAG technique modifies this pulse by adding a derivative term and an additional quadrature component, resulting in a pulse shape given by:
![]()
where:
–
is the complex envelope of the control pulse.
–
is the original Gaussian pulse.
–
is a parameter that determines the strength of the derivative correction.
The derivative term,
, acts to counteract the distortions introduced by the anharmonicity, while the imaginary component
effectively shifts the phase of the pulse, reducing the impact of the Stark shift.
Implementation and Benefits
To implement the DRAG technique, one typically starts with a standard Gaussian pulse and then calculates its derivative. The parameter
is optimized experimentally to achieve the best suppression of unwanted transitions and Stark shift effects. This optimization process involves fine-tuning the pulse parameters to match the specific characteristics of the transmon qubit being controlled.
The benefits of the DRAG technique are manifold:
1. Increased Gate Fidelity: By mitigating the effects of the Stark shift and suppressing unwanted transitions, DRAG pulses result in higher fidelity quantum gates, which are essential for reliable quantum computation.
2. Reduced Leakage: Unwanted transitions to higher energy levels (leakage) are minimized, preserving the qubit within its computational subspace (typically the |0⟩ and |1⟩ states).
3. Enhanced Robustness: DRAG pulses are more robust to variations in qubit parameters and control pulse imperfections, leading to more consistent and reliable quantum operations.
Example: Implementing DRAG Pulses
Consider a quantum gate operation, such as a single-qubit X gate (a 180-degree rotation around the X-axis of the Bloch sphere). Using a standard Gaussian pulse, the qubit might experience significant Stark shift and unwanted transitions, leading to errors. By applying the DRAG technique, the control pulse is modified to include the derivative component, which compensates for these effects.
For instance, if the original Gaussian pulse is given by:
![]()
where
is the peak amplitude and
is the pulse width, the DRAG-modified pulse becomes:
![]()
This modified pulse shape effectively reduces the Stark shift and suppresses the population of higher energy levels, resulting in a more accurate X gate operation.
Experimental Results and Practical Considerations
Experimental results have demonstrated the efficacy of the DRAG technique in various quantum computing platforms. For example, in superconducting qubit systems, implementing DRAG pulses has led to significant improvements in gate fidelities, often exceeding 99.9%. These high fidelities are important for achieving fault-tolerant quantum computation, where error rates must be below certain thresholds to enable error correction protocols.
Practical considerations for implementing DRAG pulses include the calibration of the
parameter, which may vary depending on the specific qubit and experimental setup. Additionally, the DRAG technique can be extended to more complex gate operations, such as two-qubit gates, by appropriately modifying the control pulses for each qubit involved in the operation.
Conclusion
The DRAG technique is a powerful tool in the arsenal of quantum control methods, offering a robust solution to the challenges posed by the Stark shift and unwanted transitions in transmon qubits. By carefully shaping the control pulses, DRAG enables higher fidelity quantum gates, reduced leakage, and enhanced robustness, all of which are essential for the advancement of quantum computing technologies.
Other recent questions and answers regarding Control of transmon qubits using a cryogenic CMOS integrated circuit:
- What are the challenges and potential solutions for scaling the control of transmon qubits to large-scale quantum computing systems, particularly in terms of power consumption and cooling requirements?
- What role does pulse shaping play in the control of transmon qubits, and why are Gaussian and raised cosine pulses preferred over rectangular pulses?
- How does the anharmonicity of transmon qubits aid in selective addressing of energy levels, and what are the typical frequency ranges for (omega_{01}) and (omega_{12})?
- What are the key characteristics and benefits of using transmon qubits in quantum computing, particularly in terms of their design and behavior at low temperatures?

