Values of Polynomials over Integral Domains
نویسنده
چکیده
It is well known that no nonconstant polynomial with integer coefficients can take on only prime values. We isolate the property of the integers that accounts for this, and give several examples of integral domains for which there are polynomials that only take on unit or prime values. Throughout this note, we let R denote an arbitrary UFD (unique factorization domain) that is not a field. Of course, R must be infinite. As is well known, any nonconstant polynomial with integer coefficients cannot take on only prime values. We may ask what property of the integers Z accounts for this. Here is the answer. Theorem 1. Suppose that R has only finitely many units. If f (x) ∈ R[x] is any nonconstant polynomial, then f (a) is composite for some a ∈ R. Proof. Suppose that R has k units, and let f (x) have degree d . Choose any distinct kd + 1 elements of R. Then for one of these elements b, f (b) is not a unit, as otherwise f (x) would have to take the same value more than d times, and so would be constant. Let p be any prime dividing f (b). Choose any distinct (k + 1)d + 1 elements of R congruent to b mod p. Then for one of these elements a, f (a) is neither a unit multiple of p nor is equal to 0, for the same reason, so f (a) is composite. We now want to investigate cases when we do have nonconstant polynomials that take on only prime or unit values. Definition 2. Let f (x) ∈ R[x] be a polynomial. Then f (x) is a p-polynomial if f (r) is prime for every r ∈ R, f (x) is a u-polynomial if f (r) is a unit for every r ∈ R, and f (x) is a up-polynomial if f (r) is a unit or is prime for every r ∈ R. Let P be any set of primes in Z, and let ZP denote the localization of Z at P ; concretely, ZP = {rational numbers m/n with no prime factor of n in P}. Example 3. (a) Let P = {p ≡ 3 (mod 4)} ∪ {2}, and let R = ZP . Then f (x) = x2 + 1 is an up-polynomial. (b) Let P = {p ≡ 3 (mod 4)}, and let R = ZP . Then f (x) = x2 + 1 is a upolynomial. (c) Let P = {p ≡ 5 or 7 (mod 8)} ∪ {2}, and let R = ZP . Then f (x) = (x(x + 1))2 + 2 is a p-polynomial. http://dx.doi.org/10.4169/amer.math.monthly.121.01.073 MSC: Primary 13G05 January 2014] NOTES 73 This content downloaded from 128.8.128.139 on Tue, 11 Feb 2014 10:28:35 AM All use subject to JSTOR Terms and Conditions (Parts (a) and (b) use the fact that −1 is not a quadratic residue of any prime p ≡ 3 (mod 4), and part (c) uses the fact that −2 is not a quadratic residue of any prime p ≡ 5 or 7 (mod 8).) Note that in all parts of this example, f (x) is a monic polynomial and the ring R has infinitely many distinct primes. Remark 4. We observe that if f (x) ∈ R[x] has a root in R, i.e., if f (b) = 0 for some b ∈ R, then f (x) must take on composite values. The reason is much the same as above. Choose any composite element c ∈ R. Then for any element r of R with r ≡ b (mod c), f (r) ≡ 0 (mod c), so for some a ≡ b (mod c), 0 6= f (a) ≡ 0 (mod c) is composite. We also observe that if f (x) ∈ R[x] takes on two distinct (i.e., nonassociated) prime values, then f (x) must take on composite values. For if f (b1) ≡ 0 (mod p1) and f (b2) ≡ 0 (mod p2), then by the Chinese Remainder Theorem there is a b with b ≡ b1 (mod p1) and b ≡ b2 (mod p2). For any such b, f (b) ≡ 0 (mod p1 p2), so for some a ≡ b (mod p1 p2), 0 6= f (a) ≡ 0 (mod p1 p2) is composite. Theorem 1 has a generalization, whose proof we leave to the reader. Theorem 5. Suppose that R has only finitely many units. Let Q be the quotient fieldof R. If f (x) ∈ Q[x] is any nonconstant polynomial such that f (q0) ∈ R for someq0 ∈ Q, then f (q1) ∈ R is composite for some q1 ∈ Q.Our focus in this note has been on properties of polynomials over general integraldomains. But we will remark that polynomials over the (ordinary) integers have anumber of special properties. In [1] it is shown that for any integer M , there is apolynomial of any given degree of a specific form that takes on prime values for at leastM positive integer arguments. In [3] it is shown that, assuming a standard conjecture innumber theory, for any integer M there is a quadratic polynomial of a specific form thattakes on prime values for at least M consecutive positive integer arguments. Passing topolynomials in more than one variable, [2] gives an explicit polynomial of degree 25 in26 variables whose values at integer arguments are either negative integers or (positive)primes, and which takes on all prime values. [2] also shows that any algebraic function,of any number of variables, that takes on integer values at all positive integer argumentsmust be a polynomial. (We have just sketched the main results of these papers and werefer the reader to them for more specific statements.) The proofs of these results usevery different ideas than our proofs above. REFERENCES1. B. Garrison, Polynomials with large numbers of prime values, Amer. Math. Monthly 97 (1990) 316–317,available at http://dx.doi.org/10.2307/2324515.2. J. P. Jones, D. Sato, H. Wada, D. Wiens, Diophantine representation of the set of prime numbers, Amer.Math. Monthly 83 (1976) 449–464, available at http://dx.doi.org/10.2307/2318339.3. R. A. Mollin, Prime-producing quadratics, Amer. Math. Monthly 104 (1997) 529–544, available at http://dx.doi.org/10.2307/2975080. Department of Mathematics, Lehigh University, Bethlehem, PA [email protected] 74c© THE MATHEMATICAL ASSOCIATION OF AMERICA [Monthly 121 This content downloaded from 128.8.128.139 on Tue, 11 Feb 2014 10:28:35 AMAll use subject to JSTOR Terms and Conditions
منابع مشابه
D-nice Symmetric Polynomials with Four Roots over Integral Domains D of Any Characteristic
Let D be any integral domain of any characteristic. A polynomial p(x) ∈ D[x] is D-nice if p(x) and its derivative p′(x) split in D[x]. We give a complete description of all D-nice symmetric polynomials with four roots over integral domains D of any characteristic not equal to 2 by giving an explicit formula for constructing these polynomials and by counting equivalence classes of such D-nice po...
متن کاملSolving singular integral equations by using orthogonal polynomials
In this paper, a special technique is studied by using the orthogonal Chebyshev polynomials to get approximate solutions for singular and hyper-singular integral equations of the first kind. A singular integral equation is converted to a system of algebraic equations based on using special properties of Chebyshev series. The error bounds are also stated for the regular part of approximate solut...
متن کاملModule MA3412: Integral Domains, Modules and Algebraic Integers
2 Integral Domains 12 2.1 Factorization in Integral Domains . . . . . . . . . . . . . . . . 12 2.2 Euclidean Domains . . . . . . . . . . . . . . . . . . . . . . . . 14 2.3 Principal Ideal Domains . . . . . . . . . . . . . . . . . . . . . 16 2.4 Fermat’s Two Squares Theorem . . . . . . . . . . . . . . . . . 17 2.5 Maximal Ideals and Prime Ideals . . . . . . . . . . . . . . . . 20 2.6 Unique Fact...
متن کاملNumerical solution of a class of nonlinear two-dimensional integral equations using Bernoulli polynomials
In this study, the Bernoulli polynomials are used to obtain an approximate solution of a class of nonlinear two-dimensional integral equations. To this aim, the operational matrices of integration and the product for Bernoulli polynomials are derived and utilized to reduce the considered problem to a system of nonlinear algebraic equations. Some examples are presented to illustrate the efficien...
متن کاملA solution for Volterra Integral Equations of the First Kind Based on Bernstein Polynomials
In this paper, we present a new computational method to solve Volterra integral equations of the first kind based on Bernstein polynomials. In this method, using operational matrices turn the integral equation into a system of equations. The computed operational matrices are exact and new. The comparisons show this method is acceptable. Moreover, the stability of the proposed method is studied.
متن کاملNumerical solution of nonlinear Fredholm-Volterra integral equations via Bell polynomials
In this paper, we propose and analyze an efficient matrix method based on Bell polynomials for numerically solving nonlinear Fredholm- Volterra integral equations. For this aim, first we calculate operational matrix of integration and product based on Bell polynomials. By using these matrices, nonlinear Fredholm-Volterra integral equations reduce to the system of nonlinear algebraic equations w...
متن کاملذخیره در منابع من
با ذخیره ی این منبع در منابع من، دسترسی به آن را برای استفاده های بعدی آسان تر کنید
برای دانلود متن کامل این مقاله و بیش از 32 میلیون مقاله دیگر ابتدا ثبت نام کنید
ثبت ناماگر عضو سایت هستید لطفا وارد حساب کاربری خود شوید
ورودعنوان ژورنال:
- The American Mathematical Monthly
دوره 121 شماره
صفحات -
تاریخ انتشار 2014