Primorial prime
View on WikipediaIn mathematics, a primorial prime is a prime number of the form pn# ± 1, where pn# is the primorial of pn (i.e. the product of the first n primes).[1]
Key Information
Primality tests show that:
- pn# − 1 is prime for n = 2, 3, 5, 6, 13, 24, 66, 68, 167, 287, 310, 352, 564, 590, 620, 849, 1552, 1849, 67132, 85586, 234725, 334023, 435582, 446895, ... (sequence A057704 in the OEIS). (pn = 3, 5, 11, 13, 41, 89, 317, 337, 991, 1873, 2053, 2377, 4093, 4297, 4583, 6569, 13033, 15877, 843301, 1098133, 3267113, 4778027, 6354977, 6533299, ... (sequence A006794 in the OEIS))
- pn# + 1 is prime for n = 0, 1, 2, 3, 4, 5, 11, 75, 171, 172, 384, 457, 616, 643, 1391, 1613, 2122, 2647, 2673, 4413, 13494, 31260, 33237, 304723, 365071, 436504, 498865, ... (sequence A014545 in the OEIS). (pn = 1, 2, 3, 5, 7, 11, 31, 379, 1019, 1021, 2657, 3229, 4547, 4787, 11549, 13649, 18523, 23801, 24029, 42209, 145823, 366439, 392113, 4328927, 5256037, 6369619, 7351117, 9562633, ..., (sequence A005234 in the OEIS))
The first term of the third sequence is 0 because p0# = 1 (we also let p0 = 1, see Primality of one , hence the first term of the fourth sequence is 1) is the empty product, and thus p0# + 1 = 2, which is prime. Similarly, the first term of the first sequence is not 1 (hence the first term of the second sequence is also not 2), because p1# = 2, and 2 − 1 = 1 is not prime.
The first few primorial primes are 2, 3, 5, 7, 29, 31, 211, 2309, 2311, 30029, 200560490131, 304250263527209, 23768741896345550770650537601358309 (sequence A228486 in the OEIS).
As of July 2025[ref], the largest known prime of the form pn# − 1 is 6533299# − 1 (n = 446,895) with 2,835,864 digits, found by the PrimeGrid project.
As of July 2025[update], the largest known prime of the form pn# + 1 is 9562633# + 1 (n = 637,491) with 4,151,498 digits, also found by the PrimeGrid project.
Euclid's proof of the infinitude of the prime numbers is commonly misinterpreted as defining the primorial primes, in the following manner:[2]
- Assume that the first n consecutive primes including 2 are the only primes that exist. If either pn# + 1 or pn# − 1 is a primorial prime, it means that there are larger primes than the nth prime (if neither is a prime, that also proves the infinitude of primes, but less directly; each of these two numbers has a remainder of either p − 1 or 1 when divided by any of the first n primes, and hence all its prime factors are larger than pn).
See also
[edit]References
[edit]- ^ Weisstein, Eric. "Primorial Prime". MathWorld. Wolfram. Retrieved 18 March 2015.
- ^ Michael Hardy and Catherine Woodgold, "Prime Simplicity", Mathematical Intelligencer, volume 31, number 4, fall 2009, pages 44–52.
See also
[edit]- A. Borning, "Some Results for and " Math. Comput. 26 (1972): 567–570.
- Chris Caldwell, The Top Twenty: Primorial at The Prime Pages.
- Harvey Dubner, "Factorial and Primorial Primes." J. Rec. Math. 19 (1987): 197–203.
- Paulo Ribenboim, The New Book of Prime Number Records. New York: Springer-Verlag (1989): 4.
Primorial prime
View on GrokipediaPrimorials
Definition
A primorial is a multiplicative function defined as the product of the first prime numbers, serving as the prime analog to the factorial, which multiplies the first positive integers.[2] This concept arises in number theory to capture cumulative products within the sequence of primes, highlighting their role in constructing larger integers with specific divisibility properties.[3] The standard notation for the primorial of the th prime is , denoting . For explicit computation, the first few values are (since ), , , and .[2][3] These values grow rapidly, reflecting the increasing sparsity of primes.[6] Unlike the factorial , which includes all integers from 1 to and thus incorporates composites, the primorial exclusively multiplies primes, ensuring it is square-free and divisible only by those initial primes. It also differs from other functions like the superfactorial, which involves products over factorials of primes rather than the primes themselves.[2] This distinction underscores the primorial's utility in primality testing and sieve methods.[3]Notation and examples
The standard notation for the primorial, defined as the product of the first $ n $ prime numbers, is $ p_n # $, where $ p_n $ denotes the $ n $th prime.| $ n $ | $ p_n $ | $ p_n # $ |
|---|---|---|
| 1 | 2 | 2 |
| 2 | 3 | 6 |
| 3 | 5 | 30 |
| 4 | 7 | 210 |
| 5 | 11 | 2310 |
| 6 | 13 | 30030 |
| 7 | 17 | 510510 |
| 8 | 19 | 9699690 |
| 9 | 23 | 223092870 |
| 10 | 29 | 6469693230 |
Definition of primorial primes
Formal definition
A primorial prime is a prime number $ q $ of the form $ q = p_n^# \pm 1 $, where $ p_n^# $ denotes the primorial, the product of the first $ n $ prime numbers, for some positive integer $ n $.[1] The sequence of primorial primes is cataloged as A228486 in the Online Encyclopedia of Integer Sequences (OEIS).[7] This definition encompasses both instances where $ p_n^# + 1 $ is prime and where $ p_n^# - 1 $ is prime, including rare cases where both forms yield primes for the same $ n $.[1] For the edge case $ n=1 $, where $ p_1 = 2 $ and $ 2^# = 2 $, the value $ 2^# + 1 = 3 $ is prime, whereas $ 2^# - 1 = 1 $ is not considered prime.[7][1]Variants: plus one and minus one
Primorial primes encompass two primary variants distinguished by whether one is added to or subtracted from the primorial. The general form for both is $ q = \prod_{k=1}^n p_k \pm 1 $, where $ p_k $ denotes the $ k $-th prime number.[1] The plus-one variant, $ p_n# + 1 $, frequently arises in number-theoretic constructions analogous to Euclid's classical proof of the infinitude of primes, where a number exceeding all known primes is generated to demonstrate the existence of another.[8] The indices $ n $ for which this expression yields a prime are cataloged in the On-Line Encyclopedia of Integer Sequences as A014545.[8] In contrast, the minus-one variant, $ p_n# - 1 $, These indices appear in OEIS sequence A057704.[9]Historical context
Link to Euclid's infinitude proof
Euclid's theorem, dating to approximately 300 BCE, establishes the infinitude of prime numbers through a constructive proof that assumes a finite list of primes and forms the number , which equals the primorial .[10] This , known as an Euclid number, cannot be divisible by any of the primes through , so it either is prime itself or has a prime factor larger than , thereby yielding a new prime not in the initial list.[11] Primorial primes of the form —also known as Euclid primes—arise precisely when this Euclid number is prime, representing special cases where Euclid's construction directly produces a prime rather than a composite with novel factors.[11] Such instances confirm the proof's mechanism by providing explicit examples of primes beyond any finite enumeration.[11] An extension to the construction using minus one, forming , yields another class of primorial primes, though this variant receives less emphasis in Euclid's classical argument, which relies on addition to ensure .[12] In modern number theory, primorials underpin the generation of Euclid numbers, whose prime factors—whether the number itself is prime or composite—consistently introduce primes outside the initial set, reinforcing the proof's iterative extension to infinity.[13]Early identifications and nomenclature
The construction of numbers as the product of known primes plus or minus one has ancient origins, serving as a key element in Euclid's proof of the infinitude of primes around 300 BCE, where such forms were used to generate new primes.[14] Small primorial primes, including 3, 5, 7, and 31, were recognized as primes in early number theory, with their structure as products of initial primes ±1 noted in classical and 19th-century works on prime forms, though without modern nomenclature.[1] The term "primorial" for the product of the first n primes was coined by Harvey Dubner in 1987, drawing an analogy to the factorial as a product of consecutive integers.[3] In the same publication, Dubner introduced the phrase "primorial primes" to describe primes of the form pn# ± 1, marking the formal nomenclature amid growing interest in computational prime hunting.[4] Systematic catalogs of primorial primes emerged in the 1990s through the Online Encyclopedia of Integer Sequences (OEIS), with sequences such as A018239 listing indices for pn# + 1 primes, facilitating further study.[15]Small primorial primes
List of the first few
The smallest primorial primes arise from the first few values of , where denotes the primorial (product of the first primes), along with the prime primorial 2 itself.[1] The following table enumerates these for to , listing only the cases where is prime:| Form | Prime | ||
|---|---|---|---|
| 1 | 2 | +1 | 3 |
| 2 | 6 | +1 | 7 |
| 2 | 6 | -1 | 5 |
| 3 | 30 | +1 | 31 |
| 3 | 30 | -1 | 29 |
| 4 | 210 | +1 | 211 |
| 5 | 2310 | +1 | 2311 |
| 5 | 2310 | -1 | 2309 |
| 6 | 30030 | -1 | 30029 |
| 11 | 200560490130 | +1 | 200560490131 |
| 13 | 304250263527210 | -1 | 304250263527209 |
Primality proofs for small cases
For the smallest primorial primes, such as those arising from the first few values of , primality can be verified using the basic method of trial division, which involves checking for divisibility by all prime numbers up to the square root of the candidate.[16] For instance, consider , where ; since , it suffices to test divisibility by the primes 2, 3, and 5, none of which divide 31, confirming its primality.[17] Similarly, for , the same primes up to do not divide it.[17] This approach extends straightforwardly to slightly larger small cases within . For with , , so trial division checks primes up to 13 (2, 3, 5, 7, 11, 13), revealing no divisors. For where , , requiring checks against primes up to 47; exhaustive division shows none divide 2311.[18] The case of follows analogously, with no prime divisors up to .[19] These computations are feasible by hand or simple programs due to the limited range of trial divisors. Turning to , the primorial yields , which is composite with prime factorization .[20] This factorization demonstrates non-primality directly, as both 59 and 509 are primes greater than 1. In contrast, is prime, verified by trial division against all primes up to , with no divisors found.[21] For these semi-primorial forms, where the candidate has or fully factored as the primorial (a product of known distinct primes), Pocklington's theorem provides an alternative certification method even in small cases. The theorem states that if with , , and for each prime dividing there exists an integer such that and , then is prime. Here, the complete factorization of the primorial satisfies the conditions efficiently, yielding a verifiable proof without full trial division.Mathematical properties
Connections to number theory
Primorial primes play a significant role in recursive constructions for generating new primes, such as the Euclid-Mullin sequence, where each term is the least prime factor of one plus the product of all preceding terms in the sequence. This sequence begins with 2 and proceeds by taking the least prime factor of , yielding primes like 3, 7, 43, and 13, mirroring the primorial-based approach in Euclid's proof of the infinitude of primes by ensuring each new factor exceeds previous ones.[22] Although the Euclid-Mullin sequence diverges from standard primorials after initial terms, primorial primes exemplify how products of initial primes plus or minus one can produce novel primes in such iterative processes, extending the foundational idea of Euclid's construction to analytic number theory.[1] Heuristics derived from the prime number theorem suggest that there are infinitely many primorial primes, though this remains unproven. The primorial grows exponentially, with by the prime number theorem, implying that the probability of being prime is approximately . Since the harmonic series diverges, the expected number of such primes is infinite, supporting the conjecture despite the lack of a rigorous proof. This density heuristic aligns with broader probabilistic models in number theory, where the divergence indicates positive density in the logarithmic scale, but conditional results under assumptions like the Riemann hypothesis provide only partial bounds. Large primorial primes, particularly of the form , imply the existence of substantial prime gaps immediately following them, as the subsequent integers for primes are divisible by and thus composite, creating a gap of at least . This construction demonstrates arbitrarily large gaps in the primes, with the size scaling with the primorial's magnitude, and extends to multiples of the primorial where sieving by small primes enforces extended composite runs.[23] Such implications reinforce lower bounds on maximal prime gaps in intervals up to , with primorial primes providing explicit examples of how prime density varies around highly composite numbers.[1]Large known primorial primes
Records for p_n# - 1
The largest known primorial prime of the form as of November 2025 is , with 2,835,864 digits, discovered in August 2024 through the PrimeGrid distributed computing project.[24] This prime was rigorously verified as prime using the Brillhart-Lehmer-Selfridge method with an N+1 test, confirming its status after extensive computational testing.[24] Prior to this discovery, the record was held by , which has 1,418,398 digits and was found in September 2021, also by PrimeGrid participants.[4] These advancements highlight the ongoing efforts in distributed primality searches, where multiple large candidates have emerged in recent years due to improved sieving and proving techniques. The following table summarizes the five largest known primorial primes of this form, based on digit count:| Rank | Form | Digits | Discovery Date | Project/Discoverer | |
|---|---|---|---|---|---|
| 1 | 6533299 | 2,835,864 | August 2024 | PrimeGrid (p447) | |
| 2 | 6354977 | 2,758,832 | August 2024 | PrimeGrid (p446) | |
| 3 | 4778027 | 2,073,926 | August 2024 | PrimeGrid (p442) | |
| 4 | 3267113 | 1,418,398 | September 2021 | PrimeGrid (p301) | |
| 5 | 1098133 | 476,311 | March 2012 | PrimeGrid (p346) |
Records for p_n# + 1
The largest known primorial prime of the form as of November 2025 is , which has 4,151,498 digits and was discovered on June 29, 2025, by the PrimeGrid project.[25] This prime was initially identified as a probable prime (PRP) using probabilistic tests and subsequently verified deterministically via the N-1 test with the Brillhart-Lehmer-Selfridge method, trial division, and PRP checks, confirming its primality.[25] The previous record holder was , a 3,191,401-digit prime discovered on September 20, 2024, also by PrimeGrid, which underwent similar PRP screening followed by deterministic proof using the N-1 test.[26] These discoveries highlight the ongoing efforts in distributed computing to identify large primorial primes, with PrimeGrid coordinating searches for forms like . The following table lists the five largest known primorial primes of the form , where the value given (e.g., 9562633) denotes the largest prime factor in the primorial :| Rank | Primorial Prime | Digits | Discovery Date | Discoverer (PrimeGrid Subproject) |
|---|---|---|---|---|
| 1 | 4,151,498 | June 29, 2025 | p451 | |
| 2 | 3,191,401 | September 20, 2024 | p448 | |
| 3 | 2,765,105 | August 15, 2024 | p445 | |
| 4 | 2,281,955 | July 27, 2024 | p444 | |
| 5 | 1,878,843 | April 2024 | p442 |