Take the 2-minute tour ×
Mathematics Stack Exchange is a question and answer site for people studying math at any level and professionals in related fields. It's 100% free, no registration required.

Can there be more ideals than elements of a ring? This is related to my other question Having elements as Ideals . At first glance, it seems obvious that there would be less ideals than elements of a ring. But the ideals are not necessarily mutually disjoint, and they form a subset of the power set of the ring, which is strictly larger than the ring itself. I know fields only have trivial ideals, so maybe there is an "opposite" structure that has a maximal proliferation of ideals relative to elements? A (weaker) related question is if there can be more subrings of than elements of a ring.

share|improve this question

2 Answers 2

up vote 4 down vote accepted

Start with the polynomial ring, over your favorite countable field $K$, generated by countably many indeterminates $X_q$ indexed by the rational numbers $q$. Let $R$ be the quotient of this ring obtained by imposing the equations $X_qX_r=X_q$ for all $q<r$. (In fancier language, this is the semigroup algebra over $K$ of the semigroup consisting of the rational numbers with the operation $\min$.) Then $R$ is countably infinite, but every real number $\alpha$ gives rise to an ideal $I_\alpha$ generated by all the $X_q$ with $q<\alpha$. So this countable ring has continuum many ideals.

Actually, it wasn't necessary to pass from the polynomial ring $K[X_q:q\in\mathbb Q]$ to the quotient ring $R$. $K[X_q:q\in\mathbb Q]$ itself has continuum many ideals, namely the pre-images of the $I_\alpha$'s (and lots more besides).

share|improve this answer
    
Fantastic. I will try to use this to create a ring where each element is also an ideal. –  Jacob Wakem Mar 16 at 22:04

For a finite example, consider $\mathbb{Z}[x,y,z]/(2,xx,xy,xz,yy,yz,zz)$ with 16 elements and 17 ideals.

The elements are all $a+bx+cy+dz$ for $a,b,c,d\in \{0,1\}$. The ideals are the whole ring and the 16 subspaces of $(x,y,z)$.

More generally, take $R_n = \mathbb{Z}[x_1,\ldots,x_n]/( (2) + (x_1,\ldots,x_n)^2 )$ with $2^{n+1}$ elements and $\displaystyle 1+\sum_{k=0}^n \prod_{i=1}^{n-k} \frac{2^{i+k} - 1}{2^{i\phantom{{}+k}} - 1} \approx 7.3 \cdot 2^{n^2/4}$ ideals. For example $R_4$ has 32 elements and 67 ideals, $R_5$ has 64 elements and 375 ideals, $R_6$ has 128 elements and 2826 ideals, and $R_7$ has 256 elements and 29213 ideals.

The motivation or intuition is that a vector space contain many more subspaces than elements, much like sets have many more subsets than elements. A “zero ring” is a ring in which all multiplications are 0, and so all subgroups are ideals. However, many people get confused by zero rings, so I tend to adjoin one before mentioning them. $(x_1,\cdots,x_n)/(x_1,\cdots,x_n)^2$ is a zero ring with $2^n$ elements and about $2^{n^2/4}$ ideals, and adjoining 1 (to get $R_n$) only doubles the number of elements (and gives one extra ideal).

share|improve this answer
    
Thank You. How did you come up with that? I see you both used polynomials. Just have a lot of rings you remember? –  Jacob Wakem Mar 16 at 23:46
    
Every commutative ring is a quotient of a polynomial ring. Since students study polynomials early, it is common on anonymous sites to express answers in terms of them. –  Jack Schmidt Mar 17 at 0:18
    
Great example :) –  rschwieb Mar 17 at 13:33
    
Beautiful answer. This is the math I love. –  Samuel Yusim Jul 11 at 5:55

Your Answer

 
discard

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.