The small root and the root near are regular perturbation roots: they stay finite as the parameter vanishes. Substitution of power series gives
For the third root, dominant balance for algebraic roots requires . The largest terms give , and the missing branch is . The sum of the three roots then yields
Thus the leading roots are , , and ; only the last is a divergent perturbation root. The first regular branch happens to have a zero limit, so retaining its first nonzero term is essential on the long spatial scale.
First take , the decaying half-line regime. If the three exact roots are , the exact initial value problem solution is
This follows either by solving the three initial-value equations or from the Laplace transform . The coefficient formula has sums and .
Retain the three leading roots but compute their coefficients without expanding their denominators. Set , , . A convenient composite asymptotic expansion is
It satisfies all three initial conditions exactly and retains the fast transient, the ordinary decay, and the slow decay. Its first two coefficients are and ; the fast coefficient is . Consequently the simpler bulk expression is , but that expression alone does not reproduce the initial derivative layer.
The absolute error estimate is
To see its uniformity, the slow-root error is , while its magnitude is and its coefficient is . The uniform error bound for nearby decaying exponentials therefore gives an contribution even for . The middle-root error is with coefficient , again giving . The fast-root error is with magnitude and coefficient , giving only . Coefficient errors are or smaller. This estimate concerns itself, rather than asserting the same uniform order for all derivatives or a relative error at its zero.
For the sketch, on one obtains
Thus at the origin and rises smoothly after the fast layer. For it rises towards a plateau of height approximately , then decays on the much longer scale . The bulk maximum lies near and has height asymptotic to .
Figure 1.
Initial quadratic rise, ordinary-scale plateau and slow decay of the singularly perturbed initial-value solution
. Details.
The sign of the parameter matters on an infinite interval. If is allowed, the singular mode grows rather than decays. For each fixed , its dominant contribution is . The extra in that exponent is needed for relative leading accuracy at fixed . The positive-parameter uniform absolute bound and decaying sketch do not extend to that regime.
For zero liquid flux the reduced gradients are and . Their ratio gives
Integrate to obtain the requested implicit height profile:
The endpoint relation is , and the same expression at fixes . For it simplifies to
so
This zero-liquid-flux surfactant film grows to the right from a formal dry edge, with there. Its dimensionless velocity is
The lower two thirds flow left, the upper third flows right, and the integrated liquid flux is zero. Surface advection carries the positive surfactant flux, with .
The flux curve obeys . It increases throughout the interval and reaches its formal maximum at . Larger depletion strengthens the Marangoni driving, so increasing transport is plausible. But finite flux at a zero-concentration endpoint is a singular prediction: and the height gradient diverge there. Neglected diffusion, capillarity or an endpoint region must regularize the physical limit, and can alter its maximum. The dry initial edge is likewise outside a uniform small-slope approximation.
Figure 1.
Zero-surfactant-flux and zero-liquid-flux film shapes with velocity profiles, and the reduced surfactant-flux curve versus depletion
. Details.
The profiles and velocity arrows illustrate the opposing interior-flow directions and the formal endpoint maximum; neither sketch treats the singular edges as resolved lubrication regions.
With zero surfactant flux and positive concentration, the nondiffusive surface velocity is zero. The reduced equations become , . Integrating, with the prescribed initial height and concentration, gives
At the far end, the imposed concentration drop fixes , hence
This steady surfactant film with zero surface flux rises to the right while its concentration decreases. The liquid flux is negative, despite the rightward Marangoni traction: the adverse hydrostatic gradient is strong enough to immobilize the surface and drive the interior leftwards.
In dimensionless height , the velocity profile is
It vanishes at both boundaries and is negative in the interior when . Its integral is . For , the nondegenerate film is flat and stationary. If its initial thickness is zero, the formal edge has an unbounded slope, so the lubrication approximation applies away from that edge rather than at the exact dry point.
A convenient normalization of the Papkovich–Neuber representation uses a harmonic vector and harmonic scalar :
Indeed, , so and . This is a rescaling of the usual harmonic-potential representation of Stokes flow.
Measure from the sphere center, set and . A translational vector harmonic provides the decaying force field, while the scalar dipole adjusts the surface velocity without changing that leading far field. For rotation, is harmonic, divergence-free and perpendicular to . Thus try
The translational velocity is . Matching its independent tangential and radial components at gives , ; matching rotation gives . Consequently
This superposes the translating sphere in Stokes flow and the rotating sphere in Stokes flow. It is exactly at the sphere and tends to zero at infinity; the additive ambient pressure is set to zero.
Let . Using the phase of in the singular value system, the Moore–Penrose inverse of an operator becomes
This gives the admissible data for a periodic convolution inverse. Because the range is dense and the kernel is zero here, the domain of the inverse is exactly the range, not all of .
For and ,
The formal expression would consequently be
However, the high-frequency obstruction for nonperiodic exponential data prevents this from being a Hilbert-space solution. The numerator is nonzero, is asymptotic to a nonzero constant divided by , and part (c) proved . Hence , so the Picard criterion fails. For , is not defined in the specified Hilbert space. The formal series is not a convergent generalized solution.
If with , the data are a single periodic Fourier mode: . Then there is an exact unique solution,
In particular, if the intended parameter is real, only is admissible, giving .
For the inadmissible cases the inverse problem still has approximate solutions : their images are Fourier projections converging to in , while their norms diverge. Thus the least-squares residual has infimum zero but no minimizer. This distinguishes an undefined exact inverse from a regularized truncated reconstruction; it is the necessary qualification to the question's unrestricted constant .
Use the orthonormal Fourier series basis , indexed by , and the Hilbert space inner product . The integral operator is a periodic convolution operator. Changing variables and using periodicity gives
There is no extra factor : the paper's already includes the full integral, rather than its normalized Fourier-series coefficient. The adjoint kernel is , so .
A singular system of a periodic convolution operator is therefore
Indeed and . The unit factor determined by the complex argument of in is necessary when the complex Fourier coefficients are not positive real numbers. Both families are orthonormal and complete, since every . One may enumerate by , or order the positive singular values by decreasing magnitude.
The continuous kernel on a finite square makes a Hilbert-Schmidt operator, hence compact. Since is continuously differentiable and periodic, integration by parts gives, for ,
by the Riemann-Lebesgue lemma. In particular the singular values tend to zero. Nonzero multipliers imply both and : the range is dense, but the inverse is unbounded and the range is not closed.
A singular value system of a compact operator consists of positive numbers and orthonormal families , satisfying
Thus are positive-eigenvalue eigenvectors of and of , with eigenvalue . The families are complete in and respectively. The singular values can be listed nonincreasingly with multiplicities, and tend to zero in the infinite-rank case. For finite rank there are only finitely many positive singular values; zero-kernel directions are handled separately.
Use the convention that is conjugate-linear in its first entry. Then the singular value system gives
The latter series converges precisely on the admissible range component specified by the Picard criterion:
with components annihilated by the inverse. Merely writing a formal singular expansion does not imply it converges in .
For a bounded linear operator between Hilbert spaces, let and . The Moore–Penrose inverse of an operator is defined on
For , with , define
This is independent of the chosen preimage , since two preimages differ by a vector in . Equivalently, it is the inverse of the restriction of to , applied to the component of the data in , and is zero on .
The generalized solution is the unique minimum-norm least-squares solution. Its residual is orthogonal to the range, giving the operator normal equation
The operator normal equation alone leaves an arbitrary null-space component; the second condition fixes the minimum-norm representative. The identities and hold on their appropriate domains.
For an infinite-rank compact operator, the range need not be closed, and is generally unbounded. Data outside need not have any least-squares solution at all, even though they can be approximated by range elements. This domain qualification is crucial in part (d); the Moore–Penrose notation does not turn an ill-posed inverse into an everywhere-defined bounded operator.
Let the stationary height have covariance function , with . Define the power spectrum of surface height by
For real stationary heights this spectrum is even and nonnegative. The Fourier multiplier identity in part (c) gives
Thus coherent reflection from a stationary rough surface at normal incidence is
Require the corresponding weighted spectral moment to exist. If the stationary process has a spectral measure rather than a density, the same formula uses that measure with the matching normalization.
Splitting the propagating and evanescent parts makes the effect clear:
through second order. The positive real correction reduces the magnitude of the initially negative unit coherent reflection at this order, as some reflection becomes diffuse. The evanescent part produces a coherent wave phase correction. In contrast, the first-order mean was exactly the flat reflected wave.
The height-correlation dependence of coherent reflection cannot generally be determined from alone: the quadratic term weights the whole spectrum by , while is unweighted. If the roughness varies only on scales much longer than the wavelength, so its spectrum is concentrated at , then and
This is a useful limiting formula, not the general second-order answer under only small-height assumptions. Also the mean field sampled at the moving physical boundary is through second order, from part (c), and is a different observable.
For the second-order rough-surface scattered field, the Dirichlet boundary condition expanded at the mean plane is
At normal incidence, , so its second normal derivative vanishes at zero. Define the Dirichlet-to-Neumann map for a Helmholtz half-space through the Fourier multiplier :
The first-order trace is , hence . It follows that
where with fixed regular profile. In integral notation the quadratic contribution is
The second-order field above the mean plane is added to . No local replacement of by has been made; such a replacement would be an additional long-spatial-scale approximation.
The physical surface trace and reference-plane trace are distinct. At the actual rough boundary , the condition itself says , so
The first boxed expression is the reference-plane trace needed in part (d), at , using the perturbative continuation where that plane lies below the actual boundary; the second answers the literal “at the surface” wording if it means the physical boundary. Taylor-expanding the first expression and its normal derivatives from to reproduces the second, so there is no contradiction between them.

Pinned article: Introduction to the OurBigBook Project

Welcome to the OurBigBook Project! Our goal is to create the perfect publishing platform for STEM subjects, and get university-level students to write the best free STEM tutorials ever.
Everyone is welcome to create an account and play with the site: ourbigbook.com/go/register. We belive that students themselves can write amazing tutorials, but teachers are welcome too. You can write about anything you want, it doesn't have to be STEM or even educational. Silly test content is very welcome and you won't be penalized in any way. Just keep it legal!
We have two killer features:
  1. topics: topics group articles by different users with the same title, e.g. here is the topic for the "Fundamental Theorem of Calculus" ourbigbook.com/go/topic/fundamental-theorem-of-calculus
    Articles of different users are sorted by upvote within each article page. This feature is a bit like:
    • a Wikipedia where each user can have their own version of each article
    • a Q&A website like Stack Overflow, where multiple people can give their views on a given topic, and the best ones are sorted by upvote. Except you don't need to wait for someone to ask first, and any topic goes, no matter how narrow or broad
    This feature makes it possible for readers to find better explanations of any topic created by other writers. And it allows writers to create an explanation in a place that readers might actually find it.
    Figure 1.
    Screenshot of the "Derivative" topic page
    . View it live at: ourbigbook.com/go/topic/derivative
  2. local editing: you can store all your personal knowledge base content locally in a plaintext markup format that can be edited locally and published either:
    This way you can be sure that even if OurBigBook.com were to go down one day (which we have no plans to do as it is quite cheap to host!), your content will still be perfectly readable as a static site.
    Figure 5. . You can also edit articles on the Web editor without installing anything locally.
    Video 3.
    Edit locally and publish demo
    . Source. This shows editing OurBigBook Markup and publishing it using the Visual Studio Code extension.
  3. https://raw.githubusercontent.com/ourbigbook/ourbigbook-media/master/feature/x/hilbert-space-arrow.png
  4. Infinitely deep tables of contents:
    Figure 6.
    Dynamic article tree with infinitely deep table of contents
    .
    Descendant pages can also show up as toplevel e.g.: ourbigbook.com/cirosantilli/chordate-subclade
All our software is open source and hosted at: github.com/ourbigbook/ourbigbook
Further documentation can be found at: docs.ourbigbook.com
Feel free to reach our to us for any help or suggestions: docs.ourbigbook.com/#contact