---
title: "Repulsively Bound Atom Pair Interactive Lab | Sandbox Physics"
description: "Repulsion can keep a pair together. Prepare two bosons on one lattice site. Increase their repulsion and a traveling diagonal ridge survives in the joint position map. Start the atoms apart to expose the role of preparation."
canonical_url: "https://sandboxphysics.com/labs/repulsively-bound-atom-pair/"
language: "en"
page_type: "lab"
alternate_languages:
  zh-CN: "https://sandboxphysics.com/zh/labs/repulsively-bound-atom-pair/"
last_modified: "2026-10-18"
---

QUANTUM CONFINEMENT / INTERACTION

# Repulsively Bound Atom Pair

Prepare two bosons on one lattice site. Increase their repulsion and a traveling diagonal ridge survives in the joint position map. Start the atoms apart to expose the role of preparation.

> Interactive simulation: open this page in a web browser to change the parameters and run the model.

Two identical bosons on 31 open sites in the single-band Bose–Hubbard model. A local double occupation is a superposition of bound and scattering states, not a pure bound eigenstate. Ideal number-resolved snapshots; no atom loss, higher bands, thermal bath or hardware calibration.

TWO POSITIONS · ONE STATE

ON THE SAME SITE

$0$

Model probability
Same site does not mean same place forever.

Each pillar is a pair of possible readout positions. Height: square root of joint probability; gold diagonal: same site. This is configuration space, not a physical surface.

$0$

Drag the gold launch marker, or use the arrow keys.

REPEAT. FREEZE. READ THE POSITIONS.

## Does the picture survive a measurement?

0

Every shot evolves a fresh preparation to the selected time, then measures position once. Ideal unit efficiency and number resolution. Records do not track a trajectory or reconstruct phase.

Recorded fraction

$\text{—}$

95% Wilson interval

$\text{—}$

Model center-of-mass distance

$\text{—}$

Model boundary probability

$\text{—}$

01 / THE WHOLE EVOLUTION

### Together, not motionless.

Cyan: model retained fraction. Coral: matched reference with flux or interaction removed. Gold marker: current time. Drag to inspect.

$Jt/\hbar$

02 / ONE SHOT, ONE OUTCOME

### How far apart were they?

Gold: recorded frequency. Cyan: model probability. Linear vertical scale; all shots included.

Site separation

03 / READ THE STRUCTURE

### A diagonal is a pair moving together.

Model joint probability. Coordinates are symmetrized, not labels for distinguishable atoms. Opposite cells share one unordered outcome equally. Color uses a fixed square-root gain.

04 / RAW SNAPSHOTS

### Every row starts fresh.

Latest 20 independent shots. A double mark is two atoms on one site. Full records are retained in the CSV.

Read the raw ledger

| Trial | Position | Retained |
| --- | --- | --- |

Model, units & sources

$$
\tau=Jt/\hbar,\qquad x_j=ja
$$

Static, closed, finite tight-binding systems with open ends. Time is in inverse hopping units; distance is in cell or site spacings. Both the chosen state and matched reference are evolved numerically. Edge probability warns when finite ends matter; it is not an error estimate.

$$
\frac{H}{J}=-\sum_j(b_j^\dagger b_{j+1}+\mathrm{h.c.})+\frac{U}{2J}\sum_j n_j(n_j-1)+\frac{Fa}{J}\sum_j jn_j
$$

Exactly two identical bosons on 31 sites: 496 number states, with bosonic hopping factors. Strong repulsion suppresses separation for a same-site preparation; it does not bind arbitrary initially separated atoms. The same-site fraction is not the spectral bound-state weight. Higher bands, loss, thermal noise and experimental readout errors are omitted.

$$
E_{\mathrm{bound}}(K)=\sqrt{U^2+16J^2\cos^2(Ka/2)}\quad(F=0,\ U>0)
$$

This dispersion is an infinite-chain reference, not the finite-chain solver. A nonzero slope can also limit motion; compare against the matched zero-interaction curve.

[Winkler et al. — repulsively bound pairs, Eq. 1 & Fig. 1](https://arxiv.org/html/cond-mat/0605196)
[Valiente & Petrosyan — two-particle states, Eqs. 6–8 & 18](https://arxiv.org/html/0805.1812)

Numerical evolution uses norm-limited exponential-action steps without renormalization. The diagnostic reports norm drift and energy drift. Statistical intervals describe finite ideal samples only. The 3D view and smooth maps show the model; records are independent Born-rule draws.

$0$ $0$

Physics tutorial

## Repulsively Bound Atom Pair

**Background** Prepare two bosons on one lattice site. Increase their repulsion and a traveling diagonal ridge survives in the joint position map. Start the atoms apart to expose the role of preparation.

**Why it matters** A localized picture can have very different causes. Change the cause, then measure the consequence.

Start with the essentials

- **Focus question**: Repulsion can keep a pair together.
- **One-sentence intuition**: A bounded kinetic-energy band can protect a repulsively interacting pair from separating.

### Core mathematical model

#### Unitary evolution

$$
|\psi(t)\rangle=e^{-iHt/\hbar}|\psi(0)\rangle
$$

The plotted probability and the recorded outcomes use the same state.

#### What is counted

$$
P_{\mathrm{same}}=\sum_j|\langle 2_j|\psi\rangle|^2
$$

This is a position-based event, not a measurement of phase or a bound-state projector.

### Common difficulties

#### Separate preparations

**Typical misconception** Rows show the path of one particle.

**Better mental model** Each shot starts again. The simulation is unmeasured until the selected final time.

#### Model boundary

**Typical misconception** A localized picture proves a pure bound state.

**Better mental model** Two identical bosons on 31 open sites in the single-band Bose–Hubbard model. A local double occupation is a superposition of bound and scattering states, not a pure bound eigenstate. Ideal number-resolved snapshots; no atom loss, higher bands, thermal bath or hardware calibration.

### Run the experiment

1. 01

   #### Predict

   Compare the first two presets at the same time.

   **What to observe:** Strong repulsion keeps much more probability on the diagonal.

2. 02

   #### Break the condition

   Choose the third preset.

   **What to observe:** Initially separated particles are not the same as a prepared pair.

3. 03

   #### Measure

   Collect positions, pin the result, then compare another preparation.

   **What to observe:** The estimate fluctuates, and its interval narrows with more shots.

Physical conclusion

Two identical bosons on 31 open sites in the single-band Bose–Hubbard model. A local double occupation is a superposition of bound and scattering states, not a pure bound eigenstate. Ideal number-resolved snapshots; no atom loss, higher bands, thermal bath or hardware calibration.

Quantum & fields · Starter path

## Start from a single measurement

Predict outcomes and collect counts, building from one qubit to wave packets, bound states and two-photon interference.

[Start here → **Quantum State & Born Measurement** Predict a probability, turn the analyzer, then compare it with the frequency of acquired shots. Observe · 15 min](https://sandboxphysics.com/labs/qubit-state/)

1. [1 Quantum State & Born Measurement](https://sandboxphysics.com/labs/qubit-state/)
2. [2 Stern–Gerlach Sequence](https://sandboxphysics.com/labs/stern-gerlach-sequence/)
3. [3 Wave Packet & Uncertainty](https://sandboxphysics.com/labs/wave-packet-uncertainty/)
4. [4 Quantum Harmonic Oscillator](https://sandboxphysics.com/labs/quantum-harmonic-oscillator/)
5. [5 Which Path & Quantum Eraser](https://sandboxphysics.com/labs/quantum-eraser/)
6. [6 Hong–Ou–Mandel Two-Photon Interference](https://sandboxphysics.com/labs/hong-ou-mandel/)

[View the full path →](https://sandboxphysics.com/topics/quantum/#starter-path)
