Magnetic Moments & Magnetometer theory

In classical magnetostatics, magnetic phenomena are described through the concept of the magnetic dipole. Unlike electrostatics, where isolated positive and negative electric charges exist, magnetic monopoles have never been observed in nature: Gauss's law for magnetism states that the divergence of the magnetic field vanishes everywhere: ∇ · B = 0. Consequently, all magnetic sources are fundamentally dipolar. For a macroscopic permanent bar magnet, the magnetic field can be modeled either via the Gilbert pole model or the Ampère circulating current model. In the Gilbert model, the magnet possesses two fictitious magnetic poles of equal and opposite magnetic pole strength ±p separated by an effective magnetic length 2ℓ. Due to the gradual spreading of magnetic flux near the physical ends of the bar, the effective magnetic poles are internalized, residing slightly inside the physical boundaries such that 2ℓ ≈ 0.84 × 2L (where 2L is the overall physical length). The magnetic dipole moment is defined as the vector M = p (2ℓ), directed internally from South to North. In the modern Ampèrian formulation, magnetization arises from microscopic circulating electron orbital currents and intrinsic quantum spin angular momenta: M = ∫ M_v dV. When placed in an external magnetic field B, a magnetic dipole experiences a restoring torque τ = M × B = |M| |B| sin θ, which attempts to rotate the dipole into parallel alignment with the external field. The potential energy of the dipole in the field is U = −M · B = −|M| |B| cos θ.

In 1832, Carl Friedrich Gauss derived the exact spatial magnetic field distributions for a bar magnet along its symmetry axes. In the End-on position (Tangent Position A or Tan-A), the field observation point P lies on the longitudinal axis passing through both magnetic poles. Let a bar magnet of pole strength p and semi-pole distance ℓ be centered at the origin, with its North pole at +ℓ and its South pole at −ℓ along the x-axis. The field point P is located at distance d from the center (d > ℓ). By Coulomb's inverse-square magnetic law, the field contributed by the North pole is B_N = (μ₀ / 4π) [p / (d − ℓ)²] directed away from the magnet, while the field from the South pole is B_S = (μ₀ / 4π) [p / (d + ℓ)²] directed toward the magnet. Because both vectors lie along the same line, the resultant axial magnetic field B_A is their algebraic difference: B_A = B_N − B_S = (μ₀ / 4π) p [1 / (d − ℓ)² − 1 / (d + ℓ)²] = (μ₀ / 4π) p [(d + ℓ)² − (d − ℓ)²] / [(d − ℓ)² (d + ℓ)²]. Expanding the numerator yields 4 d ℓ, so B_A = (μ₀ / 4π) [2 (p · 2ℓ) d / (d² − ℓ²)²] = (μ₀ / 4π) [2 M d / (d² − ℓ²)²]. This is the exact formula for the axial field of a finite dipole. When the observation distance d is significantly greater than the magnet half-length (d ≫ ℓ), the term (d² − ℓ²)² can be approximated as d⁴, simplifying the expression to the celebrated inverse-cube dipole law: B_A ≈ (μ₀ / 4π) (2M / d³). The magnetic field falls off inversely with the third power of distance, characteristic of all dipolar fields.

The second canonical orientation established by Gauss is the Broadside-on position (Tangent Position B or Tan-B), where the observation point P lies on the perpendicular bisector (equatorial plane) of the bar magnet at distance d from its center. In this configuration, the distances from P to both the North pole (+ℓ) and South pole (−ℓ) are identical by symmetry: r = √(d² + ℓ²). The magnitudes of the individual pole fields are equal: |B_N| = |B_S| = (μ₀ / 4π) [p / (d² + ℓ²)]. Resolving these vectors into components, the vertical components perpendicular to the magnet cancel out by anti-symmetry, while the horizontal components parallel to the magnet axis reinforce each other: B_B = 2 |B_N| cos α, where cos α = ℓ / √(d² + ℓ²). Substituting gives: B_B = 2 (μ₀ / 4π) [p / (d² + ℓ²)] [ℓ / √(d² + ℓ²)] = (μ₀ / 4π) [p (2ℓ) / (d² + ℓ²)³/²] = (μ₀ / 4π) [M / (d² + ℓ²)³/²]. For large distances (d ≫ ℓ), this simplifies to: B_B ≈ (μ₀ / 4π) (M / d³). Comparing the two Gauss configurations reveals a universal geometric theorem: at identical distances d, the magnetic field in the End-on position is exactly twice the magnetic field in the Broadside-on position: B_A = 2 B_B. In an experimental laboratory, measuring deflections in both Tan-A and Tan-B positions allows students to verify this 2:1 ratio (tan θ_A / tan θ_B ≈ 2), confirming the mathematical tensor structure of the dipolar field.

The operational metrology of the Gauss deflection magnetometer relies upon the tangent law of magnetism, which governs the equilibrium orientation of a magnetic compass needle subjected to two mutually orthogonal horizontal magnetic fields. Consider a small magnetic needle of magnetic dipole moment m_n freely pivoted at its center on an agate jewel bearing. In the absence of external magnets, Earth's horizontal geomagnetic field B_H exerts a torque that aligns the needle along the magnetic North-South meridian. In the Tan-A setup, the deflection bench is aligned strictly perpendicular to the magnetic meridian (East-West), so the axial field B produced by the test magnet at the center is strictly perpendicular to B_H. When the magnet is introduced, the needle deflects away from the meridian by an angle θ. Under this deflected condition, Earth's field exerts a restoring torque: τ_Earth = m_n B_H sin θ, attempting to pull the needle back toward the meridian. Simultaneously, the magnet's field exerts a deflecting torque: τ_magnet = m_n B cos θ, attempting to pull the needle toward the East-West axis. In static equilibrium, the two opposing torques must balance: m_n B cos θ = m_n B_H sin θ. Dividing both sides by m_n cos θ yields the tangent law: B = B_H tan θ. Substituting the inverse-cube dipole expression B = (μ₀ / 4π) (2M / d³) gives: (μ₀ / 4π) (2M / d³) = B_H tan θ, which isolates the ratio of magnetic moment to Earth's field: M / B_H = [4π / (2 μ₀)] d³ tan θ.

While a deflection magnetometer experiment directly measures the ratio of magnetic dipole moment to Earth's horizontal field (M / B_H), it cannot isolate M or B_H independently without assuming an external reference value. To solve this fundamental metrological limitation, Carl Friedrich Gauss conceived the brilliant strategy of combining deflection measurements with torsional vibration measurements. Let the same bar magnet of moment M and moment of inertia I be freely suspended horizontally in a draft-free enclosure by a torsionless unspun silk fiber in Earth's field B_H. When displaced by a small angle θ from the magnetic meridian and released, the restoring torque is τ = −M B_H sin θ ≈ −M B_H θ. The rotational equation of motion is: I d²θ/dt² + M B_H θ = 0. This is the canonical equation of simple harmonic motion with angular frequency ω = √(M B_H / I). The period of oscillation T is given by: T = 2π √(I / [M B_H]). Squaring both sides and rearranging isolates the product of M and B_H: M B_H = 4π² I / T². Having determined the ratio M / B_H = C_def from the deflection experiment and the product M B_H = C_vib from the vibration experiment, Gauss multiplied and divided the two equations: M = √(C_def × C_vib) = √[(M / B_H) (M B_H)], and B_H = √(C_vib / C_def) = √[(M B_H) / (M / B_H)]. This absolute method provided the world's first independent measurement of both Earth's magnetic field and magnetic dipole moments in absolute mechanical units without depending on any pre-calibrated electrical standards.

Precision magnetic metrology requires rigorous cancellation of mechanical, optical, and alignment artifacts. The Gauss deflection magnetometer incorporates a comprehensive eight-reading observation sequence for every distance station d. First, the compass needle carries a long, lightweight aluminum pointer mounted perpendicular to the magnetic needle. Both ends of the pointer are read on the circular protractor scale (East and West tips). Averaging the two tips cancels any eccentricity error arising from the jeweled pivot not being situated at the exact geometric center of the degree scale. Second, beneath the glass cover lies a concentric plane mirror ring. Observers align the fine pointer tip with its own reflection, eliminating sighting parallax error. Third, measurements are taken with the bar magnet placed on the East arm with its North pole pointing inward (readings 1 & 2), and then with the magnet reversed so its South pole points inward (readings 3 & 4). Reversing magnet polarity cancels any zero-offset bending of the aluminum pointer. Fourth, the magnet carrier is transferred to the West arm at the exact same distance d, and the four readings are repeated (North inward, then South inward; readings 5, 6, 7 & 8). Averaging across both arms cancels out: (1) residual leveling tilt of the bench, (2) slight zero-offset errors in the metric scale indexing, and (3) small residual angular misalignment between the bench axis and the true magnetic East-West line.

To determine the magnetic dipole moment M with statistical rigor, experimentalists vary the separation distance d across a calibrated series (e.g. 14.0 to 24.0 cm) and measure the corresponding mean deflection angle θ. From the fundamental tangent relation tan θ = [(μ₀ / 4π) (2M / B_H)] (1 / d³), designating the inverse-cube distance x = 1 / d³ (in m⁻³) as the independent variable and y = tan θ (dimensionless) as the dependent variable transforms the non-linear physical law into the canonical linear regression model: y = m x + b. Here, the regression slope m represents the physical parameter grouping: m = (μ₀ / 4π) (2M / B_H) = (2 × 10⁻⁷ M) / B_H. Inverting this relation isolates the magnetic dipole moment directly: M = (m B_H) / [2 (μ₀ / 4π)] = (m B_H) / (2 × 10⁻⁷) = m × (B_H / 2 × 10⁻⁷). With standard Earth horizontal field B_H = 3.500 × 10⁻⁵ T, the conversion factor is simply M = m × 175.0 A m². The regression intercept b serves as an indispensable experimental diagnostic: theoretically b = 0.000, so any statistically significant non-zero intercept detects residual magnetic fields from nearby ferromagnetic objects, uncompensated stray building steel, or imperfect zero-meridian alignment. Ordinary least-squares (OLS) regression minimizes the sum of squared tangent residuals, providing unbiased slope and standard error estimates.