Chapter 4: The First Qubit

4.1 What Makes Quantum Different

If you had to explain quantum computing to a friend in sixty seconds, you would need exactly three ideas. Not one, not two - all three are required, and none of them alone is sufficient. They are:

1. Superposition - A qubit can exist in a combination of 0 and 1 at the same time.
2. Measurement - When you look at a qubit, the superposition collapses to a single definite outcome.
3. Interference - Quantum states can combine so that some outcomes reinforce each other and others cancel out.

Let us build intuition for each of these in turn, starting with analogies from everyday life, and then sharpening them into precise concepts.

Superposition: Being in Two States at Once

Consider a coin spinning in the air. Before it lands, is it heads or tails? Neither - or rather, both possibilities are "alive" simultaneously. The coin has not yet committed to an outcome. A qubit in superposition is something like this: it carries the potential for both 0 and 1, with each possibility having a specific weight (called an amplitude) that determines how likely each outcome is when we finally look.

But the analogy has an important flaw. A spinning coin is actually in some definite physical state at every instant - we just do not know which side will land up. A qubit in superposition is genuinely not in the state 0 or the state 1. The superposition is not about our ignorance; it is a real, physical feature of the quantum world. This distinction matters enormously, and we will sharpen it throughout this chapter.

Measurement: The Moment of Decision

With a classical bit, you can read its value as many times as you like, and it stays the same. Measurement in the quantum world is radically different. When you measure a qubit in superposition, two things happen: (1) you get a definite result, either 0 or 1, and (2) the superposition is destroyed. The qubit is now in whatever state you observed. If you measure it again, you will get the same answer - but the original superposition is gone forever.

This is not a limitation of our instruments. It is a fundamental law of nature. There is no gentler way to extract information from a qubit. Every measurement disturbs the system. This forces quantum algorithm designers to think very carefully about when and how they measure.

Interference: The Secret Weapon

If superposition were the whole story, quantum computing would be no more powerful than flipping coins. The real power comes from interference. Think of ripples on a pond. When two ripples meet, they can either reinforce each other (constructive interference, making a bigger wave) or cancel each other out (destructive interference, flattening the water). Quantum amplitudes work the same way.

A quantum algorithm is essentially a choreography of interference. You set up a superposition over many possible answers, then apply operations that cause wrong answers to interfere destructively (canceling each other out) while correct answers interfere constructively (reinforcing each other). When you finally measure, the correct answer appears with high probability - not because the computer tried every possibility sequentially, but because the mathematics of interference concentrated the probability where you need it.

Now let us see these ideas in action. The sandbox below runs a simple quantum program: it creates a qubit, puts it in superposition with the Hadamard gate (which we will study in detail soon), and measures it 1024 times. Before you run it, take a moment to predict what you will see.

4.2 Polarized Light: Your First Quantum System

Before we dive into the abstract mathematics of qubits, let us ground our intuition in a physical system you can actually see and touch: polarized light. This is not merely an analogy - polarization of a single photon is a genuine qubit, and the experiments below demonstrate real quantum mechanics.

What Is Polarization?

Light is an electromagnetic wave. As it travels toward you, the electric field oscillates in a plane perpendicular to the direction of travel. The orientation of that oscillation is the light's polarization. If the electric field oscillates vertically, we call it vertically polarized ($\uparrow$). If it oscillates horizontally, we call it horizontally polarized ($\rightarrow$). The light can also oscillate at any angle in between, or even trace out a circle (circular polarization), but let us start with vertical and horizontal.

Experiment 1: A Single Polarizing Filter

Take a polarizing filter - the same material used in polarized sunglasses - and hold it in front of an unpolarized light source. Roughly half the light passes through. The transmitted light is now polarized along the filter's transmission axis. Rotate the filter: the brightness of the transmitted light stays the same, because unpolarized light has no preferred direction.

Here is the quantum interpretation: each photon arriving at the filter is in a superposition of "pass" and "block" relative to the filter's axis. The filter acts as a measurement device. Each photon either passes (outcome 1) or is absorbed (outcome 0). For unpolarized light, the probability is 50/50 - exactly like our Hadamard gate experiment above.

Experiment 2: Two Filters

Now place two polarizing filters in sequence. Align them both vertically: all the light that passes the first filter also passes the second. Now rotate the second filter to be horizontal (90 degrees from the first): no light gets through. This makes intuitive sense - vertically polarized light cannot pass through a horizontal filter.

In quantum language: the first filter prepares the photon in the state "vertically polarized." The second filter measures "are you horizontally polarized?" Since vertical and horizontal are orthogonal (completely different) states, the probability of passing is zero. This is the quantum version of: if a bit is definitely 0, measuring it will always give 0, never 1.

Experiment 3: The Surprise - Three Filters

Here is where things get genuinely strange. Start with two crossed filters (vertical, then horizontal) so that no light passes through. Now insert a third filter between them, oriented at 45 degrees. Classically, adding an extra barrier should block even more light. But what actually happens?

Light comes through.

This result baffled physicists when it was first demonstrated, but it follows directly from superposition and measurement. Here is the chain of events:

1. The first filter (vertical) transmits vertically polarized photons.
2. A vertically polarized photon hits the 45-degree filter. Relative to this filter's axis, the photon is in a superposition: it has a $\cos(45°) = 1/\sqrt{2}$ amplitude to pass and a $\sin(45°) = 1/\sqrt{2}$ amplitude to be blocked. So 50% of photons pass, and those that do are now polarized at 45 degrees. The measurement has changed the state.
3. A 45-degree photon hits the horizontal filter. Relative to this axis, it again has a $\cos(45°) = 1/\sqrt{2}$ amplitude to pass. So 50% of the surviving photons make it through.

The final result: 25% of the original vertically polarized photons pass through all three filters. Inserting a filter in the path actually increased the amount of light getting through from 0% to 25%. This is not a trick - it is a direct consequence of the fact that quantum measurement changes the state. The 45-degree filter did not merely observe the photon; it transformed it into a new state that now has a chance of passing the horizontal filter.

The three-filter experiment demonstrates all three of our key ideas: the photon is in a superposition relative to each filter, measurement by each filter collapses and transforms the state, and the way amplitudes combine through sequential filters is a form of interference. Keep this experiment in mind as we move to the abstract representation.

4.3 The Qubit, Geometrically

A classical bit has two states: 0 and 1. You could represent these as two points on a line. A qubit, by contrast, can exist in a continuous infinity of states - every possible superposition of $|0\rangle$ and $|1\rangle$. What shape captures all these possibilities?

The answer is a sphere: the Bloch sphere. Every single-qubit state corresponds to exactly one point on the surface of this unit sphere. It is one of the most important visualizations in all of quantum computing, and learning to read it will give you geometric intuition for everything that follows.

Reading the Bloch Sphere

The Bloch sphere uses two angles, $\theta$ (theta) and $\phi$ (phi), to specify any qubit state:

• North pole ($\theta = 0$): the state $|0\rangle$.
• South pole ($\theta = \pi$): the state $|1\rangle$.
• Equator ($\theta = \pi/2$): equal superpositions of $|0\rangle$ and $|1\rangle$ with various phases. The state $|+\rangle = \frac{1}{\sqrt{2}}(|0\rangle + |1\rangle)$ sits on the positive $x$-axis. The state $|-\rangle = \frac{1}{\sqrt{2}}(|0\rangle - |1\rangle)$ sits on the negative $x$-axis.
• $\theta$ controls the latitude: how far from the north pole (how much $|1\rangle$ is mixed in).
• $\phi$ controls the longitude: the relative phase between $|0\rangle$ and $|1\rangle$.

In mathematical terms, any single-qubit state can be written as:

Do not worry if the $e^{i\phi}$ factor looks mysterious - we will explain it in Section 4.4. For now, just notice that $\theta$ determines the measurement probabilities (the balance between $|0\rangle$ and $|1\rangle$), while $\phi$ determines the phase (a hidden angle that does not affect measurement probabilities directly, but is crucial for interference).

Interactive Bloch Sphere

Use the sliders below to explore the Bloch sphere. Adjust $\theta$ to move between the north pole ($|0\rangle$) and the south pole ($|1\rangle$). Adjust $\phi$ to rotate around the equator. Watch how the histogram changes as you vary $\theta$, and notice that $\phi$ does not affect the measurement probabilities at all - it only changes the direction on the equator.

OPENQASM 3.0;
qubit[1] q;
bit[1] c;
ry({theta}) q[0];
rz({phi}) q[0];
c = measure q;

The fact that $\phi$ does not change the measurement histogram is one of the most important subtleties of quantum mechanics. Phase is invisible when you measure in the computational basis ($|0\rangle$ vs. $|1\rangle$), but it is absolutely real - it determines how the qubit interferes with itself and with other qubits under subsequent operations. Phase is the hidden variable that makes quantum algorithms work.

4.4 The Qubit, Mathematically

We have built intuition through filters and spheres. Now it is time for the precise mathematical language. This section introduces the core formalism you will use throughout the rest of the textbook.

State Vectors

A qubit is a quantum system with two basis states, which we label $|0\rangle$ and $|1\rangle$. The general state of a qubit is a state vector:

where $\alpha$ and $\beta$ are complex numbers called amplitudes. If you have not encountered complex numbers before, here is the essential idea: a complex number has two parts, a real part and an imaginary part, written as $a + bi$ where $i = \sqrt{-1}$. Every complex number also has a magnitude (or absolute value), written $|a + bi| = \sqrt{a^2 + b^2}$.

The amplitudes $\alpha$ and $\beta$ must satisfy a single constraint called the normalization condition:

This condition ensures that the total probability of all measurement outcomes sums to 1, as it must for any valid probability distribution.

The Born Rule

When you measure a qubit in the state $|\psi\rangle = \alpha\,|0\rangle + \beta\,|1\rangle$, the outcomes and their probabilities are:

This is the Born rule, named after Max Born, who proposed it in 1926. It is one of the foundational postulates of quantum mechanics: the probability of a measurement outcome is the squared magnitude of the corresponding amplitude.

Let us see this in action with concrete examples.

Example 1: The $|0\rangle$ state. Here $\alpha = 1$ and $\beta = 0$. The probabilities are $P(0) = |1|^2 = 1$ and $P(1) = |0|^2 = 0$. Measuring always gives 0. This is a definite classical state.

Example 2: The $|+\rangle$ state. This is $|\psi\rangle = \frac{1}{\sqrt{2}}\,|0\rangle + \frac{1}{\sqrt{2}}\,|1\rangle$, where $\alpha = \beta = \frac{1}{\sqrt{2}}$. The probabilities are $P(0) = \left|\frac{1}{\sqrt{2}}\right|^2 = \frac{1}{2}$ and $P(1) = \frac{1}{2}$. This is the equal superposition we saw earlier with the Hadamard gate.

Example 3: An unequal superposition. Consider $|\psi\rangle = \frac{\sqrt{3}}{2}\,|0\rangle + \frac{1}{2}\,|1\rangle$. Check: $\left(\frac{\sqrt{3}}{2}\right)^2 + \left(\frac{1}{2}\right)^2 = \frac{3}{4} + \frac{1}{4} = 1$. Good - it is normalized. The probabilities are $P(0) = \frac{3}{4} = 75\%$ and $P(1) = \frac{1}{4} = 25\%$. This qubit is "mostly $|0\rangle$" but has a 25% chance of being found in $|1\rangle$.

Column Vector Notation

It is often convenient to represent qubit states as column vectors. We define:

Then a general qubit state becomes:

This representation makes it natural to describe quantum gates as matrices that multiply these vectors - a topic we take up in Chapter 5. For example, the Hadamard gate we have been using is the matrix:

Applying $H$ to $|0\rangle$:

This confirms what we saw in the sandbox: applying the Hadamard gate to $|0\rangle$ produces an equal superposition.

Global Phase vs. Relative Phase

An important subtlety: two state vectors that differ only by a global phase (multiplying the entire vector by some $e^{i\gamma}$) represent the same physical state. For example, $|0\rangle$ and $-|0\rangle$ and $i|0\rangle$ are all physically identical - no measurement can distinguish them.

However, a relative phase between the $|0\rangle$ and $|1\rangle$ components is physically meaningful. The states $\frac{1}{\sqrt{2}}(|0\rangle + |1\rangle)$ and $\frac{1}{\sqrt{2}}(|0\rangle - |1\rangle)$ have the same measurement probabilities in the computational basis (both give 50/50), but they are different physical states - they sit at opposite points on the Bloch sphere equator, and they behave differently under subsequent quantum gates. This is why $\phi$ mattered on the Bloch sphere even though it did not change the histogram.

4.5 Dirac Notation: The Language of Quantum Mechanics

You have already been reading expressions like $|0\rangle$ and $|\psi\rangle$ without a formal introduction. It is time to learn the notation properly. This notation, invented by the physicist Paul Dirac, is used universally in quantum computing and quantum physics. It is not merely a convention - it is designed to make quantum calculations cleaner and more intuitive.

Kets

The symbol $|\cdot\rangle$ is called a ket (rhymes with "set"). A ket represents a quantum state - a column vector in our Hilbert space. The label inside the ket is just a name:

• $|0\rangle$ - the basis state corresponding to outcome 0
• $|1\rangle$ - the basis state corresponding to outcome 1
• $|\psi\rangle$ - a general state (psi is a traditional name)
• $|+\rangle = \frac{1}{\sqrt{2}}(|0\rangle + |1\rangle)$ - the "plus" state
• $|-\rangle = \frac{1}{\sqrt{2}}(|0\rangle - |1\rangle)$ - the "minus" state

You can put any label inside a ket. Some authors write $|\uparrow\rangle$ and $|\downarrow\rangle$ for spin states, or $|H\rangle$ and $|V\rangle$ for horizontal and vertical polarization. The notation is flexible.

Bras

The bra $\langle\cdot|$ is the mirror image of the ket. Mathematically, it is the conjugate transpose (also called the Hermitian conjugate or adjoint) of the corresponding ket. If $|0\rangle = \begin{pmatrix} 1 \\ 0 \end{pmatrix}$, then $\langle 0| = \begin{pmatrix} 1 & 0 \end{pmatrix}$. If $|\psi\rangle = \begin{pmatrix} \alpha \\ \beta \end{pmatrix}$, then $\langle\psi| = \begin{pmatrix} \alpha^* & \beta^* \end{pmatrix}$, where $\alpha^*$ denotes the complex conjugate.

The naming is a pun: a "bra" and a "ket" together form a "bracket." This is deliberate.

Inner Products (Bra-Kets)

When you combine a bra and a ket, you get the inner product, written $\langle\phi|\psi\rangle$. This is a single complex number that measures the "overlap" between two quantum states. Compute it by taking the conjugate transpose of $|\phi\rangle$ and multiplying it by $|\psi\rangle$:

Key properties of the inner product:

• Orthogonality: $\langle 0|1\rangle = 0$. The basis states have zero overlap - they are completely distinguishable.
• Normalization: $\langle 0|0\rangle = 1$ and $\langle 1|1\rangle = 1$. Each basis state has unit overlap with itself.
• Born rule, restated: The probability of measuring a state $|\psi\rangle$ and obtaining outcome $|k\rangle$ is $P(k) = |\langle k|\psi\rangle|^2$.

That last point is powerful. The Born rule, which we stated earlier as "square the amplitude," is equivalent to "take the inner product with the outcome state and square its magnitude." This is the same rule, just expressed in Dirac notation. Let us verify:

Exactly as expected.

Why Dirac Notation?

You might wonder why we need special notation when column vectors work fine. The reasons become clear as systems grow:

• Compactness: Writing $|01\rangle$ is much cleaner than a 4-dimensional column vector.
• Abstraction: Dirac notation works regardless of the specific representation. You can reason about states without choosing a basis.
• Calculation: Expressions like $\langle\psi|H|\phi\rangle$ (applying a gate and taking an inner product) are easy to read and manipulate algebraically.
• Multi-qubit states: The tensor product notation $|0\rangle \otimes |1\rangle = |01\rangle$ (Chapter 6) scales naturally to any number of qubits.

4.6 Measurement: Collapsing Possibilities

We have mentioned measurement several times, but now it is time to examine it with full mathematical precision. Measurement in quantum mechanics is not a passive act of observation - it is an active physical process that fundamentally changes the state of the system.

The Measurement Postulate

When you measure a qubit in the state $|\psi\rangle = \alpha\,|0\rangle + \beta\,|1\rangle$ in the computational basis (the $|0\rangle$, $|1\rangle$ basis):

1. You obtain outcome 0 with probability $|\alpha|^2$, and the qubit's state becomes $|0\rangle$.
2. You obtain outcome 1 with probability $|\beta|^2$, and the qubit's state becomes $|1\rangle$.

The crucial point is the second part of each statement: after measurement, the qubit is in the state corresponding to the outcome. The original superposition is gone. This is sometimes called "wavefunction collapse" or "state reduction." It is irreversible - once you have measured, there is no gate you can apply to recover the original amplitudes $\alpha$ and $\beta$.

Measurement Is Irreversible

Recall from Chapter 3 that classical reversible computing preserves information. Quantum gates (which we will study in Chapter 5) are also reversible - every quantum gate has an inverse that undoes its effect. But measurement breaks this reversibility. It is the one operation in quantum mechanics that is fundamentally irreversible.

This has a practical consequence: in quantum computing, you want to delay measurement as long as possible. The longer your qubits remain in superposition, the more you can exploit interference to shape the outcome probabilities. Once you measure, the quantum advantage is gone for that qubit.

The following experiment demonstrates this irreversibility directly. We prepare a superposition, measure it, and then try to "undo" the Hadamard gate. If measurement had not collapsed the state, we would recover $|0\rangle$. But because measurement destroys the superposition, the second Hadamard operates on a definite state ($|0\rangle$ or $|1\rangle$, depending on the first measurement), and the final measurement gives a random result.

Compare this to the circuit without the intermediate measurement:

In the second circuit (no intermediate measurement), the two Hadamard gates cancel each other out ($H \cdot H = I$, the identity), so you always measure 0. In the first circuit, the intermediate measurement collapses the superposition, breaking the cancellation. This is direct experimental proof that measurement changes the quantum state.

Measurement Statistics

Because each individual measurement outcome is random, quantum computing relies on statistics. You run the same circuit many times (these repetitions are called "shots") and look at the distribution of outcomes. With enough shots, the relative frequencies converge to the true probabilities predicted by the Born rule.

How many shots do you need? It depends on the precision you want. With $N$ shots, the statistical uncertainty in an estimated probability $p$ is approximately $\sqrt{p(1-p)/N}$. For 1024 shots and $p = 0.5$, this gives an uncertainty of about $\pm 1.6\%$ - good enough for most demonstrations, but real quantum experiments often use tens of thousands of shots.

The following Predict-Observe-Explain exercise tests your understanding of measurement and state collapse.

No-Cloning and the Limits of Measurement

One more profound consequence of the measurement postulate: you cannot determine an unknown quantum state from a single copy. If someone gives you a single qubit in an unknown state $|\psi\rangle = \alpha\,|0\rangle + \beta\,|1\rangle$, a single measurement gives you only one bit of information (either 0 or 1), and it destroys the state in the process. You cannot go back and measure again, because the state has changed.

Furthermore, the no-cloning theorem (which we will prove in Chapter 7) tells you that there is no way to create an exact copy of an unknown quantum state. This means you cannot make many copies and measure each one to build up statistics. A single unknown qubit state is fundamentally unknowable from one copy.

This is not a bug - it is a feature. The no-cloning theorem is what makes quantum cryptography possible: an eavesdropper cannot copy quantum information without disturbing it, so any interception is detectable.

Review