Jump to content

Pythagorean prime: Difference between revisions

From Wikipedia, the free encyclopedia
Content deleted Content added
2 is prime, and is the sum of two squares, but is not Pythagorean
 
(47 intermediate revisions by 18 users not shown)
Line 1: Line 1:
{{redirect|Pythagorean number|the field invariant related to sums of squares|Pythagoras number|elements of extension fields containing square roots of sums of squares|Pythagorean field}}
{{Refimprove|date=November 2007}}
[[File:Squared right triangle.svg|thumb|360px|The Pythagorean prime 5 and its square root are both hypotenuses of [[right triangle]]s with integer legs. The formulas show how to transform any right triangle with integer legs into another right triangle with integer legs whose hypotenuse is the square of the first triangle's hypotenuse.]]
{{redirect|Pythagorean number|the field invariant related to sums of squares|Pythagoras number}}
A '''Pythagorean prime''' is a [[prime number]] of the form 4''n'' + 1. Pythagorean primes are exactly the odd prime numbers that are the sum of two squares (and from this derives the name in reference to the famous [[pythagorean theorem]].)
A '''Pythagorean prime''' is a [[prime number]] of the {{nowrap|form <math>4n+1</math>.}} Pythagorean primes are exactly the odd prime numbers that are the sum of two squares; this characterization is [[Fermat's theorem on sums of two squares]].


Equivalently, by the [[Pythagorean theorem]], they are the odd prime numbers <math>p</math> for which <math>\sqrt p</math> is the length of the [[hypotenuse]] of a [[right triangle]] with integer legs, and they are also the prime numbers <math>p</math> for which <math>p</math> itself is the hypotenuse of a primitive [[Pythagorean triangle]]. For instance, the number 5 is a Pythagorean prime; <math>\sqrt5</math> is the hypotenuse of a right triangle with legs 1 and 2, and 5 itself is the hypotenuse of a right triangle with legs 3 and 4.

==Values and density==
The first few Pythagorean primes are
The first few Pythagorean primes are
:[[5 (number)|5]], [[13 (number)|13]], [[17 (number)|17]], [[29 (number)|29]], [[37 (number)|37]], [[41 (number)|41]], [[53 (number)|53]], [[61 (number)|61]], [[73 (number)|73]], [[89 (number)|89]], [[97 (number)|97]], [[101 (number)|101]], [[109 (number)|109]], [[113 (number)|113]], {{OEIS|id=A002144}}.
{{bi|left=1.6|[[5 (number)|5]], [[13 (number)|13]], [[17 (number)|17]], [[29 (number)|29]], [[37 (number)|37]], [[41 (number)|41]], [[53 (number)|53]], [[61 (number)|61]], [[73 (number)|73]], [[89 (number)|89]], [[97 (number)|97]], [[101 (number)|101]], [[109 (number)|109]], [[113 (number)|113]], ... {{OEIS|id=A002144}}.}}
By [[Dirichlet's theorem on arithmetic progressions]], this sequence is infinite. More strongly, for each <math>n</math>, the numbers of Pythagorean and non-Pythagorean primes up to <math>n</math> are approximately equal. However, the number of Pythagorean primes up to <math>n</math> is frequently somewhat smaller than the number of non-Pythagorean primes; this phenomenon is known as {{nowrap|[[Chebyshev's bias]].{{r|rubsar}}}} For example, the only values of <math>n</math> up to 600000 for which there are more Pythagorean than non-Pythagorean odd primes less than or equal to n are 26861 {{nowrap|and 26862.{{r|gramar}}}}


==Representation as a sum of two squares==
[[Fermat's theorem on sums of two squares]] states that these primes can be represented as sums of two squares uniquely (up to order), and that no other primes can be represented this way, aside from 2=1<sup>2</sup>+1<sup>2</sup>. Thus these primes (and 2) occur as norms of [[Gaussian integers]], while other primes do not.
The sum of one odd square and one even square is congruent to 1&nbsp;mod&nbsp;4, but there exist [[composite number]]s such as 21 that are {{nowrap|1 mod 4}} and yet cannot be represented as sums of two squares. [[Fermat's theorem on sums of two squares]] states that the [[prime number]]s that can be represented as sums of two squares are exactly 2 and the odd primes congruent to {{nowrap|1 mod 4.{{r|stewart}}}} The representation of each such number is unique, up to the ordering of the two squares.{{r|leveque}}


By using the [[Pythagorean theorem]], this representation can be interpreted geometrically: the Pythagorean primes are exactly the odd prime numbers <math>p</math> such that there exists a [[right triangle]], with integer legs, whose [[hypotenuse]] has {{nowrap|length <math>\sqrt p</math>.}} They are also exactly the prime numbers <math>p</math> such that there exists a right triangle with integer sides whose hypotenuse has {{nowrap|length <math>p</math>.}} For, if the triangle with legs <math>x</math> and <math>y</math> has hypotenuse length <math>\sqrt p</math> (with <math>x>y</math>), then the triangle with legs <math>x^2-y^2</math> and <math>2xy</math> has hypotenuse {{nowrap|length <math>p</math>.{{r|stillwell}}}}
The law of [[quadratic reciprocity]] says that if ''p'' and ''q'' are distinct odd primes, at least one of which is Pythagorean, then
''p'' is a [[quadratic residue]] mod ''q'' [[if and only if]] ''q'' is a quadratic residue mod ''p''; by contrast, if neither ''p'' nor ''q'' is Pythagorean, then ''p'' is a quadratic residue mod ''q'' if and only if ''q'' is '''not''' a quadratic residue mod ''p''.


Another way to understand this representation as a sum of two squares involves [[Gaussian integer]]s, the [[complex number]]s whose real part and imaginary part are both {{nowrap|integers.{{r|mazur}}}} The norm of a Gaussian integer <math>x+iy</math> is the {{nowrap|number <math>x^2+y^2</math>.}} Thus, the Pythagorean primes (and 2) occur as norms of Gaussian integers, while other primes do not. Within the Gaussian integers, the Pythagorean primes are not considered to be prime numbers, because they can be factored as
In the field Z/p with p a Pythagorean prime, the polynomial x^2 = -1 has two solutions.
<math display=block>p=(x+iy)(x-iy).</math>
Similarly, their squares can be factored in a different way than their [[integer factorization]], as
<math display=block>
\begin{align}
p^2&=(x+iy)^2(x-iy)^2\\
&=(x^2-y^2+2ixy)(x^2-y^2-2ixy).\\
\end{align}</math>
The real and imaginary parts of the factors in these factorizations are the leg lengths of the right triangles having the given hypotenuses.

==Quadratic residues==
The law of [[quadratic reciprocity]] says that if <math>p</math> and <math>q</math> are distinct odd primes, at least one of which is Pythagorean, then <math>p</math> is a [[quadratic residue]] {{nowrap|mod <math>q</math>}} [[if and only if]] <math>q</math> is a quadratic residue {{nowrap|mod <math>p</math>;}} by contrast, if neither <math>p</math> nor <math>q</math> is Pythagorean, then <math>p</math> is a quadratic residue {{nowrap|mod <math>q</math>}} if and only if <math>q</math> is '''not''' a quadratic residue {{nowrap|mod <math>p</math>.{{r|leveque}}}}

In the [[finite field]] <math>\Z/p</math> with <math>p</math> a Pythagorean prime, the polynomial equation <math>x^2=-1</math> has two solutions. This may be expressed by saying that <math>-1</math> is a quadratic residue {{nowrap|mod <math>p</math>.}} In contrast, this equation has no solution in the finite fields <math>\Z/p</math> where <math>p</math> is an odd prime but is not {{nowrap|Pythagorean.{{r|leveque}}}}

[[File:Paley13.svg|thumb|The Paley graph with 13 vertices]]
For every Pythagorean prime <math>p</math>, there exists a [[Paley graph]] with <math>p</math> vertices, representing the numbers {{nowrap|modulo <math>p</math>,}} with two numbers adjacent in the graph if and only if their difference is a quadratic residue. This definition produces the same adjacency relation regardless of the order in which the two numbers are subtracted to compute their difference, because of the property of Pythagorean primes that <math>-1</math> is a quadratic {{nowrap|residue.{{r|chung}}}}

==References==
{{reflist|refs=

<ref name=chung>{{citation
| last = Chung | first = Fan R. K. | author-link = Fan Chung
| isbn = 9780821889367
| pages = 97–98
| publisher = American Mathematical Society
| series = CBMS Regional Conference Series
| title = Spectral Graph Theory
| url = https://books.google.com/books?id=YUc38_MCuhAC&pg=PA97
| volume = 92
| year = 1997}}</ref>

<ref name=gramar>{{citation
| last1 = Granville | first1 = Andrew | author1-link = Andrew Granville
| last2 = Martin | first2 = Greg
| date=January 2006
| doi = 10.2307/27641834
| title = Prime number races
| journal = [[The American Mathematical Monthly]]
| volume = 113 | issue = 1 | pages = 1--33
| url = http://www.dms.umontreal.ca/%7Eandrew/PDF/PrimeRace.pdf
| jstor = 27641834}}</ref>

<ref name=leveque>{{citation
| last = LeVeque | first = William Judson | author-link = William J. LeVeque
| isbn = 9780486689067
| pages = 100, 103, 183
| publisher = Dover
| title = Fundamentals of Number Theory
| url = https://books.google.com/books?id=F6aJtNcwyw8C&pg=PA183
| year = 1996}}</ref>

<ref name=mazur>{{citation
| last = Mazur | first = Barry | author-link = Barry Mazur
| editor-last = Gowers | editor-first = Timothy | editor-link = Timothy Gowers
| contribution = Algebraic numbers [IV.I]
| isbn = 9781400830398
| pages = 315–332
| publisher = Princeton University Press
| title = [[The Princeton Companion to Mathematics]]
| year = 2010}} See in particular section 9, "Representations of Prime Numbers by Binary Quadratic Forms", [https://books.google.com/books?id=ZOfUsvemJDMC&pg=PA325 p.&nbsp;325].</ref>

<ref name=rubsar>{{citation
| last1 = Rubinstein | first1 = Michael
| last2 = Sarnak | first2 = Peter
| doi = 10.1080/10586458.1994.10504289
| issue = 3
| journal = [[Experimental Mathematics (journal)|Experimental Mathematics]]
| pages = 173–197
| title = Chebyshev's bias
| volume = 3
| year = 1994}}</ref>

<ref name=stewart>{{citation
| last = Stewart | first = Ian | author-link = Ian Stewart (mathematician)
| isbn = 9780465082377
| page = 264
| publisher = Basic Books
| title = Why Beauty is Truth: A History of Symmetry
| url = https://books.google.com/books?id=6akF1v7Ds3MC&pg=PA264
| year = 2008}}</ref>

<ref name=stillwell>{{citation
| last = Stillwell | first = John | author-link = John Stillwell
| isbn = 9780387955872
| page = 112
| publisher = Springer
| series = [[Undergraduate Texts in Mathematics]]
| title = Elements of Number Theory
| url = https://books.google.com/books?id=LiAlZO2ntKAC&pg=PA112
| year = 2003}}</ref>

}}

==External links==
* {{citation|last=Eaves|first=Laurence|title=Pythagorean Primes: including 5, 13 and 137|url=http://www.numberphile.com/videos/pythagorean_primes.html|work=Numberphile|publisher=[[Brady Haran]]|authorlink=Laurence Eaves|access-date=2013-04-02|archive-url=https://web.archive.org/web/20160319164414/http://www.numberphile.com/videos/pythagorean_primes.html|archive-date=2016-03-19|url-status=dead}}
* {{OEIS el|A007350|name=Where prime race 4n-1 vs. 4n+1 changes leader}}


[[Category:Classes of prime numbers]]
[[Category:Classes of prime numbers]]
[[Category:Eponymous numbers in mathematics]]
[[Category:Squares in number theory]]


{{Prime number classes|state=collapsed}}
{{Prime number classes|state=collapsed}}

{{numtheory-stub}}

Latest revision as of 19:09, 22 October 2023

The Pythagorean prime 5 and its square root are both hypotenuses of right triangles with integer legs. The formulas show how to transform any right triangle with integer legs into another right triangle with integer legs whose hypotenuse is the square of the first triangle's hypotenuse.

A Pythagorean prime is a prime number of the form . Pythagorean primes are exactly the odd prime numbers that are the sum of two squares; this characterization is Fermat's theorem on sums of two squares.

Equivalently, by the Pythagorean theorem, they are the odd prime numbers for which is the length of the hypotenuse of a right triangle with integer legs, and they are also the prime numbers for which itself is the hypotenuse of a primitive Pythagorean triangle. For instance, the number 5 is a Pythagorean prime; is the hypotenuse of a right triangle with legs 1 and 2, and 5 itself is the hypotenuse of a right triangle with legs 3 and 4.

Values and density

[edit]

The first few Pythagorean primes are

5, 13, 17, 29, 37, 41, 53, 61, 73, 89, 97, 101, 109, 113, ... (sequence A002144 in the OEIS).

By Dirichlet's theorem on arithmetic progressions, this sequence is infinite. More strongly, for each , the numbers of Pythagorean and non-Pythagorean primes up to are approximately equal. However, the number of Pythagorean primes up to is frequently somewhat smaller than the number of non-Pythagorean primes; this phenomenon is known as Chebyshev's bias.[1] For example, the only values of up to 600000 for which there are more Pythagorean than non-Pythagorean odd primes less than or equal to n are 26861 and 26862.[2]

Representation as a sum of two squares

[edit]

The sum of one odd square and one even square is congruent to 1 mod 4, but there exist composite numbers such as 21 that are 1 mod 4 and yet cannot be represented as sums of two squares. Fermat's theorem on sums of two squares states that the prime numbers that can be represented as sums of two squares are exactly 2 and the odd primes congruent to 1 mod 4.[3] The representation of each such number is unique, up to the ordering of the two squares.[4]

By using the Pythagorean theorem, this representation can be interpreted geometrically: the Pythagorean primes are exactly the odd prime numbers such that there exists a right triangle, with integer legs, whose hypotenuse has length . They are also exactly the prime numbers such that there exists a right triangle with integer sides whose hypotenuse has length . For, if the triangle with legs and has hypotenuse length (with ), then the triangle with legs and has hypotenuse length .[5]

Another way to understand this representation as a sum of two squares involves Gaussian integers, the complex numbers whose real part and imaginary part are both integers.[6] The norm of a Gaussian integer is the number . Thus, the Pythagorean primes (and 2) occur as norms of Gaussian integers, while other primes do not. Within the Gaussian integers, the Pythagorean primes are not considered to be prime numbers, because they can be factored as Similarly, their squares can be factored in a different way than their integer factorization, as The real and imaginary parts of the factors in these factorizations are the leg lengths of the right triangles having the given hypotenuses.

Quadratic residues

[edit]

The law of quadratic reciprocity says that if and are distinct odd primes, at least one of which is Pythagorean, then is a quadratic residue mod if and only if is a quadratic residue mod ; by contrast, if neither nor is Pythagorean, then is a quadratic residue mod if and only if is not a quadratic residue mod .[4]

In the finite field with a Pythagorean prime, the polynomial equation has two solutions. This may be expressed by saying that is a quadratic residue mod . In contrast, this equation has no solution in the finite fields where is an odd prime but is not Pythagorean.[4]

The Paley graph with 13 vertices

For every Pythagorean prime , there exists a Paley graph with vertices, representing the numbers modulo , with two numbers adjacent in the graph if and only if their difference is a quadratic residue. This definition produces the same adjacency relation regardless of the order in which the two numbers are subtracted to compute their difference, because of the property of Pythagorean primes that is a quadratic residue.[7]

References

[edit]
  1. ^ Rubinstein, Michael; Sarnak, Peter (1994), "Chebyshev's bias", Experimental Mathematics, 3 (3): 173–197, doi:10.1080/10586458.1994.10504289
  2. ^ Granville, Andrew; Martin, Greg (January 2006), "Prime number races" (PDF), The American Mathematical Monthly, 113 (1): 1--33, doi:10.2307/27641834, JSTOR 27641834
  3. ^ Stewart, Ian (2008), Why Beauty is Truth: A History of Symmetry, Basic Books, p. 264, ISBN 9780465082377
  4. ^ a b c LeVeque, William Judson (1996), Fundamentals of Number Theory, Dover, pp. 100, 103, 183, ISBN 9780486689067
  5. ^ Stillwell, John (2003), Elements of Number Theory, Undergraduate Texts in Mathematics, Springer, p. 112, ISBN 9780387955872
  6. ^ Mazur, Barry (2010), "Algebraic numbers [IV.I]", in Gowers, Timothy (ed.), The Princeton Companion to Mathematics, Princeton University Press, pp. 315–332, ISBN 9781400830398 See in particular section 9, "Representations of Prime Numbers by Binary Quadratic Forms", p. 325.
  7. ^ Chung, Fan R. K. (1997), Spectral Graph Theory, CBMS Regional Conference Series, vol. 92, American Mathematical Society, pp. 97–98, ISBN 9780821889367
[edit]