An exact differential equation is a first-order ordinary differential equation of the form M(x,y)dx + N(x,y)dy = 0 where the mixed partial derivatives of M and N are equal: ∂M/∂y = ∂N/∂x. This condition ensures the existence of a potential function ψ(x,y) such that dψ = M dx + N dy.
The core idea behind exactness lies in vector calculus: if a differential form represents the gradient of a scalar field, then its integral along any path depends only on endpoints, not the route taken. For the equation to be exact, this path independence must hold globally within the domain of interest.
Exact differential equations arise naturally in physics and engineering. The work done by conservative forces, such as gravity or electrostatic fields, follows this structure. When the force field is conservative, the work integral becomes path-independent - a property mirrored mathematically by exactness.
A differential equation M(x,y)dx + N(x,y)dy = 0 is exact if there exists a continuously differentiable function ψ(x,y) such that:
If such a function ψ exists, then the solution to the differential equation is given implicitly by ψ(x,y) = C, where C is a constant.
To verify exactness, compute ∂M/∂y and ∂N/∂x. If they are equal over a simply connected region, the equation is exact. This test is known as the exactness condition or cross-partial test.
For example, consider the equation:
(2xy + 3)dx + (x² - 1)dy = 0Let M = 2xy + 3 and N = x² - 1. Compute:
Since both partials are equal, the equation is exact.
There are two primary cases: exact equations and non-exact ones. Only exact equations allow direct integration via a potential function. Non-exact equations require transformation using an integrating factor.
When an equation fails the exactness test, it may still become exact after multiplication by a suitable integrating factor μ(x,y). Finding μ is often non-trivial and requires solving another PDE. However, if μ depends only on x or only on y, the process simplifies.
| Type | Condition | Method |
|---|---|---|
| Exact | ∂M/∂y = ∂N/∂x | Integrate M w.r.t. x, N w.r.t. y, combine results |
| Non-exact (μ depends on x) | ∂M/∂y ≠ ∂N/∂x | Solve dμ/dx = μ·(∂M/∂y − ∂N/∂x)/N |
| Non-exact (μ depends on y) | ∂M/∂y ≠ ∂N/∂x | Solve dμ/dy = μ·(∂N/∂x − ∂M/∂y)/M |
Some equations are exact only in restricted domains. For instance, the equation (−y/x²)dx + (1/x)dy = 0 is exact everywhere except at x = 0. The singularity at x = 0 breaks the domain's simple connectivity, invalidating global exactness even if local conditions hold.
Many assume that all first-order ODEs can be solved by exact methods. That's false. Only those satisfying the cross-partial equality qualify. Attempting to apply exact techniques to non-exact equations leads to incorrect solutions.
Another error is treating exactness as a property of the equation alone, ignoring the domain. A function pair M,N may satisfy ∂M/∂y = ∂N/∂x locally but fail globally due to discontinuities or singularities. The domain must be simply connected for the converse of the exactness theorem to hold.
Also, people often confuse exactness with separability. While some exact equations are separable, most are not. Separation of variables applies only when M/N depends solely on x or y. Exactness is a broader, more general condition.
Consider a particle moving under a central force field described by F = −k/r² in polar coordinates. The radial component of motion yields the equation:
(r³ cos θ) dr + (r³ sin θ) dθ = 0Set M = r³ cos θ, N = r³ sin θ. Then:
These are unequal, so the equation is not exact. But multiplying through by μ(r) = 1/r³ gives:
(cos θ) dr + (sin θ) dθ = 0Now M = cos θ, N = sin θ. Check:
Still not equal. Try μ(r,θ) = 1/r². After testing, the correct integrating factor turns out to be μ = 1/r². Multiplying yields:
(cos θ / r) dr + (sin θ / r) dθ = 0Now ∂M/∂θ = −sin θ / r, ∂N/∂r = −sin θ / r. They match. The equation is now exact. Integrate M w.r.t. r: ∫ cos θ / r dr = cos θ ln|r| + h(θ). Differentiate w.r.t. θ: −sin θ ln|r| + h'(θ). Set equal to N: sin θ / r → contradiction unless scaled. Eventually, ψ(r,θ) = sin θ ln|r| + C.
I once spent two hours debugging a model of heat diffusion in a rod because I assumed the boundary condition equation was exact. It wasn't. The mismatch in partials revealed a missing term in the flux law. Fixing the physical model restored exactness and gave a valid analytical solution.
Exact differential equations are solvable when their components meet the cross-partial condition. This condition guarantees the existence of a potential function whose level sets define the solution curves. Not all equations are exact - and assuming they are leads to errors. Always check ∂M/∂y = ∂N/∂x before proceeding.
Understanding what is exact differential equation enables deeper insight into conservation laws, thermodynamics, and fluid dynamics. The method extends beyond pure math into applied sciences where conservation principles govern behavior.