Quantum Mechanics in Computing refers to the application of the principles of quantum mechanics to information processing, computation and the physical representation of information. Classical computing represents information using bits with states conventionally labelled 0 and 1, whereas quantum computing represents information using quantum states whose behaviour follows quantum-mechanical rules. The relevant principles include superposition, measurement, unitary evolution, interference, entanglement, quantisation and the probabilistic nature of measurement outcomes.
Quantum computation does not simply mean performing conventional calculations with smaller electronic components; it uses a different mathematical and physical model of computation. A quantum system can evolve according to a controlled quantum circuit and produce a probability distribution over possible measurement results. The computational objective is therefore generally to engineer interference so that amplitudes associated with useful answers are enhanced and amplitudes associated with undesirable answers are suppressed.
Superposition is one of the most important principles applied in quantum computing. A single Quantum Bit can exist in a state represented mathematically as a linear combination of basis states, usually written as alpha times the state 0 plus beta times the state 1, subject to a normalisation condition on the corresponding probabilities. This does not mean that measurement reveals two ordinary classical values simultaneously.
When measured in the computational basis, the result is either 0 or 1 with probabilities determined by the state's amplitudes. The ability to manipulate amplitudes before measurement is computationally important because quantum operations can cause amplitudes to interfere constructively or destructively. The distinction between probability and probability amplitude is fundamental: amplitudes can be positive, negative or complex and can interfere, whereas ordinary classical probabilities cannot reproduce this behaviour directly.
Entanglement provides another major quantum-mechanical resource used in computing. When multiple quantum systems become entangled, their joint state cannot generally be represented as independent states for each component. Measurements on one component can therefore exhibit correlations with another component that cannot be described by assigning each system an independent classical state in advance. In quantum computing, entanglement is generated through multi-qubit operations such as controlled gates and is used within many algorithms, error-correction codes and computational protocols.
However, entanglement alone does not constitute a computational speed-up. Useful quantum computation requires carefully controlled evolution and measurement, and some quantum circuits can generate highly entangled states without providing a practical advantage over classical methods. Entanglement also does not violate relativistic causality or provide a mechanism for transmitting usable information instantaneously.
Quantum measurement introduces a fundamental difference from classical computation. Before measurement, a quantum state can contain coherent amplitudes for several possible outcomes, but a measurement in a chosen basis produces a classical result according to the quantum measurement probabilities. The act of measurement generally changes the state and can destroy information about phase relationships that existed beforehand. Consequently, quantum algorithms cannot normally be inspected step by step in the same way as classical variables without disturbing the computation.
Quantum circuits are designed so that measurement occurs after quantum operations have transformed the state in a useful way. Different measurement bases can reveal different information about a state, and advanced protocols may use ancillary qubits and repeated measurements to extract particular properties without directly reading every component of an underlying quantum state.
Unitary evolution is another central principle. In an ideal closed quantum computation, gates correspond to unitary transformations that preserve the total norm of the quantum state and are reversible. Common single-qubit operations include the Pauli X, Y and Z operations, the Hadamard operation and phase rotations, while multi-qubit operations include controlled-NOT and controlled-phase operations. Reversibility distinguishes ideal quantum gates from many ordinary classical logic operations, such as irreversible AND gates, although reversible classical computation is also possible.
Practical quantum hardware introduces noise, decoherence and measurement errors, so physical quantum computers are not perfectly isolated mathematical systems. Quantum error correction addresses this gap by encoding logical information across multiple physical quantum systems and detecting or correcting errors without directly measuring the encoded logical state.
Quantum interference is what turns superposition into a computational mechanism rather than merely a statistical mixture. A quantum algorithm manipulates amplitudes so that computational paths combine with relative phases that affect the probability of later measurement outcomes. Shor's algorithm, for example, uses quantum operations and the quantum Fourier transform to extract periodicity information relevant to integer factorisation, while Grover's algorithm uses amplitude amplification to increase the probability of finding a marked item in an unstructured search space.
The resulting speed-up is not obtained by literally trying every candidate and reading all answers at once. Instead, the algorithm creates a structured interference pattern from which useful information can be extracted through measurement. This distinction is essential to understanding why quantum computation differs from simplistic descriptions of quantum parallelism.
Quantum mechanics also imposes fundamental constraints on computing. The no-cloning theorem states that an unknown arbitrary quantum state cannot be copied perfectly, which affects information storage, communication and error correction. Quantum states are also susceptible to decoherence and noise, making physical implementation substantially more demanding than abstract circuit notation suggests.
The amount of classical information required to describe a general quantum state grows exponentially with the number of qubits, although this does not mean a quantum computer automatically possesses an exponential computational advantage. Classical simulation can sometimes exploit structure in quantum states, while tensor-network methods, stabiliser methods and specialised simulation techniques can efficiently represent important classes of quantum computations. Consequently, the computational value of quantum mechanics depends on the interaction between physical resources, algorithmic structure, complexity theory and the specific problem being solved.
The relationship between Quantum Mechanics in Computing and classical computer science is therefore complementary rather than purely competitive. Quantum computers use classical concepts such as algorithms, complexity classes, programming languages, compilers, optimisation, error decoding and numerical analysis, while their underlying computational model introduces quantum states and operations. Quantum complexity theory studies classes such as BQP, which characterises problems efficiently solvable with bounded-error quantum computation under standard definitions.
Quantum algorithms can provide polynomial or superpolynomial improvements for particular problem families, but not every classical task becomes faster. The field's central technical challenge is to identify computational problems where quantum mechanical resources can be transformed into meaningful algorithmic advantage and then implement the required operations with sufficiently low error and sufficient scale to outperform practical classical alternatives.
