## Abstract

Let \(R=k[t^{n_{1}},\ldots ,t^{n_{s}}]=k[x_{1},\ldots ,x_{s}]/P\) be a numerical semigroup ring and let *P*^{(n)} = *P*^{n}*R*_{P} ∩ *R* be the symbolic power of *P* and *R*_{s}(*P*) = ⊕_{i≥ 0}*P*^{(n)}*t*^{n} the symbolic Rees ring of *P*. It is hard to determine symbolic powers of *P*; there are even non-Noetherian symbolic Rees rings for 3-generated semigroups. We determine the primary decomposition of powers of *P* for some classes of 3-generated numerical semigroups.

## Introduction

If *I* is an ideal in a Noetherian ring *R*, then the subring *R*(*I*) = *R*[*I**t*] of *R*[*t*] is called the Rees ring of *I*. This was introduced by Rees, in his proof of the Artin-Rees lemma. If *P* is a prime ideal, then the primary decomposition of *P*^{n} always contains a *P*-primary component, it is \(R=k[t^{n_1},\ldots ,t^{n_s}]=k[x_1,\ldots ,x_n]/P\) and is called the symbolic *n* th power of *P*. Cowsik [1] asked if the symbolic Rees algebra *R*_{s}(*P*) = *R*[*P**t*,*P*^{(2)}*t*^{2},*P*^{(3)}*t*^{3},…] is always Noetherian. This was shown not to be true by Roberts [7, 8]. There are even counterexamples when \(R=k[t^{n_{1}},t^{n_{2}},t^{n_{3}}]=k[x,y,z]/P\). Goto-Nishida-Watanabe [2] showed that for *n* ≥ 4 then *k*[*t*^{7n− 3},*t*^{(5n− 2)n},*t*^{8n− 3}] does not have a finitely generated symbolic Rees algebra if char(*k*) = 0. Their smallest counterexample is *k*[*t*^{25},*t*^{72},*t*^{29}] = *k*[*x*,*y*,*z*]/(*x*^{11} − *y**z*^{7},*y*^{3} − *x*^{4}*z*^{4},*z*^{11} − *x*^{7}*y*^{2}). Hochster [4] has shown that *P*^{(n)} = *P*^{n} for all *n* if *k*[*x*_{1},…,*x*_{s}]/*P* is a complete intersection. Huneke [5] has shown that if *P* is a prime ideal of height 2 in a 3-dimensional ring, then the opposite holds.

## Numerical Semigroup Rings

If \(R=k[t^{n_{1}},\ldots ,t^{n_{s}}]\), we map *k*[*x*_{1},…,*x*_{s}] to *k*[*t*], by \(x_{i}\mapsto t^{n_{i}}\). Then *R* ≃ *k*[*x*_{1},…,*x*_{s}]/*P*, and *P* is a prime ideal since *R* is a domain. In the case of numerical semigroup rings, *R* is 1-dimensional, so *P*^{(n)} = *P*^{n} or *P*^{(n)} ∩ *Q*, where *Q* is (*x*_{1},…,*x*_{s})-primary. If *P*^{(n)} = *P*^{n} for all *n*, i.e., if the semigroup ring is a complete intersection, then *R*_{s}(*P*) = *R*(*P*) is Noetherian and *P*^{n} is *P*-primary. Kunz [6] has shown that the semigroup ring of *S* is Gorenstein if and only if the semigroup *S* is symmetric, i.e., if for some *D*, we have *n* ∈ *S* if and only if *D* − *n*∉*S*.

## 3-Generated Numerical Semigroups

In the sequel, we mean numerical semigroup when we write semigroup. A 3-generated semigroup is Gorenstein if and only if it is a complete intersection. This is so, since if we factor with a generator, we get a ring of embedding dimension 2. The factor ring is Gorenstein (complete intersection) if and only if the semigroup ring is Gorenstein (complete intersection). For a ring of embedding dimension 2, the concepts are equivalent. If the semigroup is generated by 3 elements, and is not a complete intersection, then \(R=k[t^{n_{1}},t^{n_{2}},t^{n_{3}}]\simeq k[x,y,z]/P\) where *P* is generated by the three 2 × 2-minors of a matrix (the relation matrix)

Herzog and Ulrich [3] have shown that *R*_{s}(*P*) = *R*[*P**t*,*P*^{(2)}*t*^{2}] if and only if *a*_{1} = *a*_{2},*b*_{1} ≤ *b*_{2},*c*_{1} ≥ *c*_{2} (or a permutation). Huneke [5] has shown that if *P* is a 2-dimensional prime in a 3-dimensional ring, then *P*^{(2)}/*P*^{2} is generated by one element Δ. Schenzel [9] has, in the case of a 3-generated semigroup, determined Δ. The result is, if *a*_{1} ≤ *a*_{2},*b*_{1} ≥ *b*_{2},*c*_{1} ≥ *c*_{2}, in particular this holds if *R*_{s}(*P*) = *R*[*P**t*,*P*^{(2)}*t*^{2}] according to the result of Herzog and Ulrich,

He also showed that \((x^{a_{1}},y^{b_{2}},z^{c_{2}}){\Delta }\in P^{2}\).

In the other case, *a*_{1} ≥ *a*_{2}, *b*_{1} ≥ *b*_{2}, *c*_{1} ≥ *c*_{2}, there is a similar result:

and \((x^{a_{2}},y^{b_{2}},z^{c_{2}}){\Delta }\in P^{2}\). If *R*_{s}(*P*)≠*R*[*P**t*,*P*^{(2)}*t*^{2}], we can only determine the primary decomposition of *P*^{2}, but if *R*_{s}(*P*) = *R*[*P**t*,*P*^{(2)}*t*^{2}], we will construct the primary decomposition of *P*^{n} for all *n* in the following cases: if the semigroup is generated by an arithmetic sequence, or if it is generated by *a* < *b* < *c* with *c* − *a* ≤ 4, or if the multiplicity is ≤ 4.

###
**Theorem 1**

*Suppose that**R* = *k*[*t*^{a},*t*^{b},*t*^{c}] = *k*[*x*,*y*,*z*]/*P**is not a complete intersection. Then*\(P^{2} = (({\Delta }) + P^{2}) \cap ((z^{c_{2}}) + P^{2})\)*is**a primary decomposition. If furthermore**R*_{s}(*P*) = *R*[*P**t*,*P*^{(2)}*t*^{2}]*,**then*\(P^{2n} = (P^{(2)})^{n}\cap ((z^{nc_{2}}) + P^{2n})\)*and*\(P^{2n + 1} = P(P^{(2)})^{n}\cap ((z^{nc_{2}}) + P^{2n})\)*.*

### Proof

Since *P*^{(2)} = (Δ) + *P*^{2} and since \((z^{c_{2}}) + P^{2}\) is (*x*,*y*,*z*)-primary, it suffices to note that \((z^{c_{2}})\cap ({\Delta })=z^{c_{2}} {\Delta }\subseteq P^{2}\) ([9, Theorem 10.3]) to see the first statement. If *R*_{s}(*P*) = *R*[*P**t*,*P*^{(2)}*t*^{2}], then \(P^{(2n)} ={\sum }_{i = 0}^{n}(P^{(2)})^{i}P^{2n-2i} = (P^{(2)})^{n}\) since *P*^{2} ⊆ *P*^{(2)}. In the same way, we see that *P*^{(2n+ 1)} = *P**P*^{(2n)}. Finally, \((z^{nc_{2}})\cap P^{(2n)} =(z^{nc_{2}})\cap (({\Delta })+(P^{2})^{n}) \subseteq P^{2n}\) since \(z^{c_{2}}{\Delta }\subseteq P^{2}\).

The remaining part is a search for examples when *R*_{s}(*P*) = *R*[*P**t*,*P*^{(2)}*t*^{2}]. □

### Arithmetic Sequences

Now suppose that the semigroup is generated by *m*,*m* + *d*,*m* + 2*d*, \(\gcd (m,m+d,m + 2d)= 1\). The semigroup is symmetric (so the semigroup ring is a complete intersection) if *m* is even and *d* odd. Otherwise, the relation matrix is

Thus, according to the theorem by Herzog and Ulrich *R*_{s}(*P*) = *R*[*P**t*,*P*^{(2)}*t*^{2}]. We use this result also below to see when *R*_{s}(*P*) = *R*[*P**t*,*P*^{(2)}*t*^{2}].

###
**Theorem 2**

*Let**R* = *k*[*t*^{m},*t*^{m+d},*t*^{m+ 2d}] = *k*[*x*,*y*,*z*]/*P**be nonsymmetric. Then the primary decomposition is**P*^{2n} = ((Δ) + *P*^{2})^{n} ∩ ((*z*^{n}) + *P*^{2})^{n}*and**P*^{2n+ 1} = (*P*(Δ) + *P*^{2})^{n} ∩ ((*z*^{n}) + *P*^{2})^{n}*.*

### Semigroups Generated by *a* < *b* < *c*, *c* − *a* ≤ 4

If the semigroup is not generated by an arithmetic sequence, the generators are *m*,*m* + 1,*m* + 3 or *m*,*m* + 1,*m* + 4 or *m*,*m* + 2,*m* + 3 or *m*,*m* + 3,*m* + 4.

If the semigroup is generated by *m*,*m* + 1,*m* + 3, it is symmetric if *m* ≡ 0 (mod 3).

If *m* = 3*k* + 1 the relation matrix is

and *R*_{s}(*P*)≠*R*[*P**t*,*P*^{(2)}*t*^{2}] for all *k*.

If *m* = 3*k* + 2 the relation matrix is

and *R*_{s}(*P*)≠*R*[*P**t*,*P*^{(2)}*t*^{2}] unless *k* = 1.

If the semigroup is generated by *m*,*m* + 1,*m* + 4, it is symmetric if *m* ≡ 0 (mod 4) (and if *m* = 5).

If *m* = 4*k* + 1, *k* ≥ 2, the relation matrix is

and *R*_{s}(*P*)≠*R*[*P**t*,*P*^{(2)}*t*^{2}].

If *m* = 4*k* + 2, *k* ≥ 2, the relation matrix is

and *R*_{s}(*P*) = *R*[*P**t*,*P*^{(2)}*t*^{2}] if and only if *k* ≥ 3 and *k* = 1.

If *m* = 4*k* + 3, the relation matrix is

and *R*_{s}(*P*) = *R*[*P**t*,*P*^{(2)}*t*^{2}] only if *k* = 1 or *k* = 2.

If the semigroup is generated by *m*,*m* + 2,*m* + 3, it is symmetric if *m* ≡ 0 (mod 3) (and if *m* = 4).

If *m* = 3*k* + 1, the relation matrix is

and *R*_{s}(*P*)≠*R*[*P**t*,*P*^{(2)}*t*^{2}] for all *k*.

If *m* = 3*k* + 2 the relation matrix is

and *R*_{s}(*P*)≠*R*[*P**t*,*P*^{(2)}*t*^{2}] for all *k*.

If the semigroup is generated by *m*,*m* + 3,*m* + 4, it is symmetric if *m* ≡ 0 (mod 4) (and if *m* = 6 or *m* = 9).

If *m* = 4*k* + 1, *k* ≥ 2, the relation matrix is

and *R*_{s}(*P*)≠*R*[*P**t*,*P*^{(2)}*t*^{2}].

If *m* = 4*k* + 2, *k* ≥ 2, the relation matrix is

and *R*_{s}(*P*) = *R*[*P**t*,*P*^{(2)}*t*^{2}] if and only if *k* ≥ 4.

If *m* = 4*k* + 3 the relation matrix is

and *R*_{s}(*P*) = *R*[*P**t*,*P*^{(2)}*t*^{2}] only if *k* = 3.

For *m* = 5 the relation matrix is

and *R*_{s}(*P*) = *R*[*P**t*,*P*^{(2)}*t*^{2}].

###
**Theorem 3**

*If the semigroup is generated by**a* < *b* < *c**,**c* − *a* ≤ 4*,**not symmetric, and**a*,*b*,*c**not an arithmetic sequence, then**R*_{s}(*P*) = *R*[*P**t*,*P*^{(2)}*t*^{2}] *if and only if the generators are* 5,6,8 *or* 6,7,10 *or* 15,18,19 *or* 7,8,11 *or* 11,12,15 *or* 15,18,19 *or* 4*k* + 2,4*k* + 3,4*k* + 6*,**k* ≥ 3 *or* 4*k* + 2,4*k* + 5,4*k* + 6*,**k* ≥ 4*.*

### Semigroups of Multiplicity 3

We note that if the semigroup is of multiplicity ≤ 4, then Huneke [5] has shown that the symbolic Rees ring is Noetherian. Suppose that the semigroup is generated by 3,3*k* + 1,3*l* + 2. In order to have a 3-generated semigroup, we must have *l* ≤ 2*k* and *k* ≤ 2*l* + 1. The semigroup is never symmetric. The relation matrix is

and *R*_{s}(*P*) = *R*[*P**t*,*P*^{(2)}*t*^{2}].

### Semigroups of Multiplicity 4

If a 3-generated semigroup has multiplicity 4 and is not symmeteric, it has generators 4,4*k* + 1,4*l* + 3. If *k* > *l* the relation matrix is

and *R*_{s}(*P*) = *R*[*P**t*,*P*^{(2)}*t*^{2}] if and only if 5*l* − 3*k* + 3 ≤ 0. If *k* ≤ *l*, the relation matrix is

and *R*_{s}(*P*) = *R*[*P**t*,*P*^{(2)}*t*^{2}] if and only if 3*l* − 5*k* + 1 ≥ 0.

###
**Theorem 4**

*If the semigroup has multiplicity* 3*,**then**R*_{s}(*P*) = *R*[*P**t*,*P*^{(2)}*t*^{2}]*.**If the multiplicity is* 4 *and not symmetric, it is generated by* 4,4*k* + 1,4*l* + 3*,**and**R*_{s}(*P*) = *R*[*P**t*,*P*^{(2)}*t*^{2}] *if and only if**k* > *l**and* 5*l* − 3*k* + 3 ≤ 0 *or if**k* ≤ *l**and* 3*l* − 5*k* + 1 ≥ 0*.*

## References

- 1.
Cowsik, R.: Symbolic powers and the number of defining equations. Algebra and its Applications, Lecture Notes in Pure and Appl. Math. 9, 13–14 Dekker New York (1984)

- 2.
Goto, S., Nishida, K., Watanabe, K.: Non-Cohen-Macaulay symbolic blow-ups for space monomial curves and counterexamples to Cowsik’s question. Proc. Am. Math. Soc.

**120**(2), 383–392 (1994) - 3.
Herzog, J., Ulrich, B.: Self-linked curve singularities. Nagoya Mat. J.

**120**, 129–153 (1990) - 4.
Hochster, M.: Criteria for equality of ordinary and symbolic powers of primes. Math. Z.

**133**, 53–65 (1973) - 5.
Huneke, C.: Primary components and integral closures of ideals in 3-dimensional regular local rings. Math. Ann.

**275**, 617–635 (1986).*c*.*i*. - 6.
Kunz, E.: Almost complete intersections are not Gorenstein rings. J. Algebra

**28**, 111–115 (1974) - 7.
Roberts, P.: A prime ideal in a polynomial ring whose symbolic blow-up is not Noetherian. Proc. Am. Math. Soc.

**94**, 589–592 (1985) - 8.
Roberts, P.: An infinitely generated symbolic blow-up in a power series ring and a new counterexample to Hilbert’s fourteenth problem. J. Algebra

**132**, 461–473 (1990) - 9.
Schenzel, P.: Examples of Noetherian symbolic blow-up rings. Rev. Roumaine Math. Pures Appl.

**33**, 375–383 (1988)

## Author information

### Affiliations

### Corresponding author

## Additional information

*To Le Tuan Hoa on his sixtieth birthday*

### Publisher’s Note

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

## Rights and permissions

**OpenAccess** This article is distributed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made.

## About this article

### Cite this article

Fröberg, R. Primary Decomposition of Powers of the Prime Ideal of a Numerical Semigroup Ring.
*Acta Math Vietnam* **44, **665–669 (2019). https://doi.org/10.1007/s40306-018-00321-2

Received:

Revised:

Accepted:

Published:

Issue Date:

### Keywords

- Primary decomposition
- Symbolic power
- Numerical semigroup ring

### Mathematics Subject Classification (2010)

- Primary 13A30
- Secondary 13H10