Which order do you take derivative of delta function?

Click For Summary
SUMMARY

The discussion centers on the order of taking derivatives of the delta function in double integrals, specifically the expression I=∫∫dxdy [∇x∇y δ(x-y)] f(x)g(y). Participants analyze the implications of switching the order of derivatives, demonstrating that while the calculations yield different intermediate forms, the final results remain consistent. The conversation highlights the importance of careful notation and the use of integration by parts in resolving such expressions.

PREREQUISITES
  • Understanding of delta functions in calculus
  • Familiarity with partial derivatives and notation (∇x, ∇y)
  • Knowledge of integration by parts
  • Basic principles of functional analysis
NEXT STEPS
  • Study the properties of the delta function in distribution theory
  • Learn advanced techniques in integration by parts for multiple integrals
  • Explore the implications of switching derivative orders in functional analysis
  • Investigate the application of delta functions in physics and engineering contexts
USEFUL FOR

Mathematicians, physicists, and engineers who work with differential equations and need to understand the behavior of delta functions in integrals.

geoduck
Messages
257
Reaction score
2
If you have I=∫∫dxdy [∇xy δ(x-y)] f(x)g(y)

where ∇x is the derivative with respect to x (and similarly for y), then doesn't it matter which order you take the derivatives? For example:

I=∫∫dxdy f(x) ∇x [∇y δ(x-y)] g(y)
=∫dx f(x) ∇x[-g'(x)]=∫dx f(x) [-g''(x)]

whereas if you take the other order:

I=∫∫dxdy ∇y [∇x δ(x-y)] f(x)g(y)
=∫∫dxdy g(y)∇y [∇x δ(x-y)] f(x)
=∫dx g(x) [-f''(x)]
 
Physics news on Phys.org
I don't follow your calculation. Assuming notation isn't misleading, and writing things mathematics-style rather than physics-style, the calculation would go

[tex] \begin{align*}<br /> \int_{\mathbb{R}} \int_{\mathbb{R}} (\nabla_x \nabla_y \delta(x-y)) f(x) g(y) \, dx \, dy<br /> &= - \int_{\mathbb{R}} \int_{\mathbb{R}} (\nabla_y \delta(x-y)) \nabla_x(f(x) g(y)) \, dx \, dy<br /> \\& = - \int_{\mathbb{R}} \int_{\mathbb{R}} (\nabla_y \delta(x-y)) \nabla_x(f(x) g(y)) \, dy \, dx<br /> \\& = \int_{\mathbb{R}} \int_{\mathbb{R}} \delta(x-y) \nabla_y \nabla_x(f(x) g(y)) \, dy \, dx<br /> \\& = \cdots<br /> \end{align*}[/tex]

Of course, I think the end result is the same -- just continue using integration by parts to switch between the result you get from this and the two results you got.

EDIT: I see how your calculation works now: if the inner integral is in terms of y, you are factoring the [itex]\nabla_x[/itex] outside of the inner integral.
 

Similar threads

  • · Replies 4 ·
Replies
4
Views
2K
  • · Replies 19 ·
Replies
19
Views
4K
  • · Replies 1 ·
Replies
1
Views
2K
  • · Replies 5 ·
Replies
5
Views
3K
  • · Replies 11 ·
Replies
11
Views
2K
  • · Replies 6 ·
Replies
6
Views
4K
  • · Replies 1 ·
Replies
1
Views
2K
Replies
1
Views
2K
Replies
0
Views
2K
  • · Replies 0 ·
Replies
0
Views
1K