Mathematics Stack Exchange is a question and answer site for people studying math at any level and professionals in related fields. Join them; it only takes a minute:

Sign up
Here's how it works:
  1. Anybody can ask a question
  2. Anybody can answer
  3. The best answers are voted up and rise to the top

Let $f$ be a real polynomial of two variables. Let $\partial_f=f\left(\frac{\partial}{\partial x},\frac{\partial}{\partial y}\right)$. Let $H$ denote the space of harmonic polynomials, i.e., polynomials $h$ such that $\partial_{x^2+y^2}h=0$.

Is it true that if the kernel of $\partial_f$ contains $H$ then $x^2+y^2|f(x,y)$?

I would appreciate an elementary proof/disproof.

share|cite|improve this question
Maybe this might be interesting... – draks ... Mar 30 '12 at 19:34
up vote 4 down vote accepted

It seems the answer is yes. We can write $F(x,y)=\sum_{d=0}^Nf_d(x,y)$ where $f_d$ is a homogeneous polynomial of degree $d$. Since for all $n$, $P_n(x,y):=(x+iy)^n+(x-iy)^n$ and $Q_n(x,y):=(x+iy)^n-(x-iy)^n$ are harmonic (because holomorphic), we should have $\partial_FP_n=\partial_FQ_n=0$. If $d>n$, we have $\partial_{f_d}P_n=\partial_{f_d}Q_n=0$. We have to show that for all $2\leq d\leq N$, $(x^2+y^2)\mid f_d(x,y)$ (we have $f_0=f_1=0$, evaluating $\partial_F$ on $1$, $x$ and $y$). We fix $d$ and we assume that $(x^2+y^2)\mid f_k$ for $2\leq k< d$. We have $\partial_{f_d}P_d=\partial_{f_d}Q_d=0$. We can write $f_d=\sum_{k=0}^da_{d,k}x^ky^{d-k}$. So \begin{align*} \partial_{f_d}P_d&=\sum_{k=0}^da_{d,k}\partial_x^k\partial_y^{d-k}\left((x+iy)^d+(x-iy)^d\right)\\ &=\sum_{k=0}^da_{d,k}\frac{d!}{(d-k)!}\partial_y^{d-k}\left((x+iy)^{d-k}+(x-iy)^{d-k}\right)\\ &=\sum_{k=0}^da_{d,k}\frac{d!}{(d-k)!}\left((i^{d-k}+(-i)^{d-k}\right)(d-k)!\\ &=\sum_{k=0}^da_{d,k}d!\left((i^{d-k}+(-i)^{d-k}\right) \end{align*} hence $$\tag{1}\sum_{k=0}^da_{d,k}i^{d-k}(1+(-1)^{d-k})=0\mbox{ and }\sum_{k=0}^da_{d,k}i^{d-k}(1-(-1)^{d-k})=0.$$ Now, we have to consider the case $d$ odd and $d$ even. If $d=2p$, then we have relationships of the form $$a_{2p,2p}-a_{2p,2p-2}+a_{2p,2p-2}-\cdots= 0$$ and $$a_{2p,2p-1}-a_{2p,2p-3}+a_{2p,2p-5}-\cdots= 0.$$ Putting that into the expression of $f_{2p}$, we get after computations $$f_{2p}(x,y)=\sum_{k=1}^pa_{2p,2k}y^{2d-2k}(x^{2k}-(-1)^ky^{2k}) +\sum_{k=1}^{p-1}a_{2p,2k+1}y^{2d-2k-1}x(x^{2k}-(-1)^ky^{2k}),$$ and since $x^{2k}-(-1)^ky^{2k}=(x^2)^k-(-y^2)^k=(x^2+y^2)\sum_{l=0}^{k-1}x^{2l}(-y^2)^{l-k}$, $(x^2+y^2)\mid f_{2p}$.

The same computations give the result for $d$ odd, since if $d=2p+1$, (1) becomes $$\sum_{k=0}^{2p+1}a_{2p+1,k}i^{-k}(1-(-1)^k)=0\mbox{ and }\sum_{k=0}^{2p+1}a_{2p+1,k}i^{-k}(1+(-1)^k)=0.$$

share|cite|improve this answer
Two questions now: what are the polynomials $P_0\in \mathbb R[x,y]$ such that for all $f$, if $\partial_f (P)=0$ for all $P\in \ker\partial_{P_0}$, then $P_0\mid P$? And what about other dimensions? – Davide Giraudo Dec 24 '11 at 16:42

Your Answer


By posting your answer, you agree to the privacy policy and terms of service.

Not the answer you're looking for? Browse other questions tagged or ask your own question.