Vector Calculus for Symmetry (IPhO)

Vector calculus becomes easy when you stop treating it as algebra and start treating it as geometry: flux, circulation, and symmetry. In IPhO problems, the “calculus” is often one line once the surface/loop is chosen well.

1. Definitions (Must Know)

A. Flux and circulation

• Flux of a vector field vector F through a surface S:

  Φ = ∬_S vector F · d vector A,

  where d vector A = n hat dA points along the chosen normal (often outward).

• Circulation of vector F around a closed loop C:

  Γ = ∮_C vector F · d vector ℓ,

  where d vector ℓ is tangent to the loop in the chosen direction.

B. Gradient, divergence, curl (what they measure)

• Gradient: ∇ φ points in the direction of steepest increase of scalar field φ.
• Divergence: ∇ · vector F measures net “outflow” per unit volume (sources/sinks).
• Curl: ∇ × vector F measures local tendency to circulate (swirl/rotation).

C. Theorems that turn calculus into geometry

• Divergence theorem:

  ∬_(∂ V) vector F · d vector A = ∭_V(∇ · vector F) dV.

• Stokes’ theorem:

  ∮_(∂ S) vector F · d vector ℓ = ∬_S(∇ × vector F) · d vector A.

D. Gauss and Ampere in physics form

• Gauss’s law (electrostatics):

  ∬ vector E · d vector A = Q_enc/ε₀.

• Ampère’s law (magnetostatics):

  ∮ vector B · d vector ℓ = μ₀ I_enc.

E. “Gaussian surface” and “Amperian loop”

• A Gaussian surface is the closed surface you choose to evaluate ∬ vector E · d vector A.
• An Amperian loop is the closed loop you choose to evaluate ∮ vector B · d vector ℓ.

They are not physical objects; they are geometry choices that must respect the symmetry of the field.

2. Key Ideas (What Earns Marks)

• Use symmetry to argue:
  • the direction of the field (which components must be zero),
  • what the field can depend on (often only r),
  • where the magnitude is constant on a chosen surface/loop.
• Choose S or C so the integral collapses:
  • vector F · d vector A is constant on a surface patch, or is zero on most of the surface.
  • vector F · d vector ℓ is constant and tangential along the loop, or zero along the parts that remain.
• State the “constant-on-the-surface” sentence explicitly. This is where method marks live.
• Use enclosed quantity correctly:
  • Q_enc is only charge inside the Gaussian surface.
  • I_enc is only current that pierces the surface bounded by your loop.
• Check units and limiting behaviour (e.g. far-field scaling, behaviour at the symmetry axis).

3. Detailed Explanations

A. The symmetry test (when Gauss/Ampere will actually solve)

Gauss/Ampere give you one scalar equation:

• Gauss: ∬ vector E · d vector A = Q_enc/ε₀.
• Ampere: ∮ vector B · d vector ℓ = μ₀ I_enc.

They “solve” the field only if symmetry lets you take vector E or vector B outside the integral as a constant magnitude.

If symmetry is weak, the laws are still true, but they become constraints, not direct solution tools.

B. How to write a symmetry argument (use this template)

In many marking schemes, this sequence earns marks even before the integral:

1. “By [spherical/cylindrical/planar] symmetry, vector F must point along [radial/normal/tangential] direction.”
2. “The magnitude | vector F| can depend only on [the radius r / distance to the plane / etc.].”
3. “Choose a [sphere/cylinder/pillbox/circle] where | vector F| is constant on the relevant part of the surface/loop.”
4. Evaluate the integral and solve for | vector F|.

C. High-yield Gaussian surfaces (electrostatics)

• Spherical symmetry (point charge, spherically symmetric charge density):

  • use a sphere of radius r.
  • vector E is radial and constant in magnitude on the sphere.
  • flux becomes Φ = E(4π r²).

• Cylindrical symmetry (infinite line charge, long cylinder):

  • use a cylinder of radius r and length L.
  • vector E is radial; the curved surface contributes, the end caps often contribute zero by geometry.
  • flux becomes Φ = E(2π rL).

• Planar symmetry (infinite sheet):

  • use a pillbox that straddles the sheet.
  • vector E is normal to the sheet and constant on each cap.
  • flux becomes Φ = EA + EA = 2EA.

D. High-yield Amperian loops (magnetostatics)

• Long straight wire: circle of radius r around the wire.

  • vector B is tangential and constant on the circle.
  • ∮ vector B · d vector ℓ = B(2π r).

• Long solenoid: rectangle loop with one long side inside the solenoid.

  • inside field approximately uniform and parallel to axis; outside approximately small.
  • the integral reduces to Bℓ (plus negligible contributions).

• Toroid: circle inside the core (field tangential; depends on r).

E. Divergence and curl as quick diagnostics

• If ∇ · vector F = 0 in a region, there is no net source/sink in that region (flux through any closed surface in that region is zero).
• If ∇ × vector F = 0 in a simply connected region, circulation around any contractible loop is zero and the field is conservative there.

In IPhO settings, you often use these as:

• a check on a derived field (does it have sources only where the charge/current is?),
• a fast way to evaluate an integral via theorems when ∇ · vector F or ∇ × vector F is simple.

4. Common Mistakes

• Choosing a Gaussian surface or loop that breaks symmetry, then wrongly pulling | vector F| outside the integral.
• Forgetting that vector F · d vector A and vector F · d vector ℓ are dot products (direction matters).
• Using Q_enc or I_enc incorrectly (including sources that lie outside the chosen surface).
• Missing piecewise results when the source distribution changes with radius.
• Not stating why some parts of the surface/loop contribute zero.

5. Exam Tips

• Draw the symmetry and the chosen surface/loop. Even a rough sketch clarifies the dot products.
• Write one sentence that pins the direction: “radial”, “normal”, or “tangential”.
• Before integrating, say where the field is constant and where contributions vanish.
• After you solve for E(r) or B(r), check:
  • units,
  • scaling with r,
  • limiting cases (e.g. r → ∞, r → 0 if appropriate).

6. Worked Examples

A. Infinite line charge (Gauss)

An infinite line has uniform charge per unit length λ. Find E(r) at distance r from the line.

B. Uniformly charged solid sphere (Gauss, piecewise)

A solid sphere of radius R has uniform volume charge density ρ. Find E(r).

C. Long straight wire (Ampere)

A long straight wire carries current I. Find B(r) at distance r from the wire.

D. Using Stokes when the curl is simple

In some region, a field satisfies (∇ × vector F) = k z hat (constant). Find ∮_C vector F · d vector ℓ where C is a circle of radius a in the xy-plane oriented counterclockwise (viewed from + z hat).

7. Mind Stretchers

A. Infinite sheet (pillbox geometry)

An infinite sheet has surface charge density σ. Use Gauss’s law to find the magnitude of vector E on either side.

B. Coaxial cable (piecewise Ampere)

A long coaxial cable has current + I uniformly distributed in a solid inner conductor of radius a, and return current -I uniformly distributed in a thin outer cylindrical shell at radius b (assume negligible thickness). Find B(r) in the regions r < a, a < r < b, and r > b.

C. Curl-free but not “obviously conservative”

Consider vector F = k/r φ hat in the plane (cylindrical polar coordinates). What is ∮ vector F · d vector ℓ around a circle of radius a centred at the origin?

8. Practice