The factory of prime numbers
Ancient Origins
The ancient Greeks, driven by a passion for knowledge and mathematical beauty, gave the world its first formal glimpse into the realm of prime numbers. Euclid, through his legendary "Elements," provided definitions and theorems that remain foundational to this day.
Yet, whispers of primes can be traced back even earlier, in the ancient scrolls of Egypt, Babylon, and China, where the seeds of mathematical discovery were first sown. The journey of prime numbers is a tale of human curiosity, brilliance, and the eternal quest to understand the fabric of our numerical universe.
Basic Concepts
We consider only positive integers: 1, 2, 3, and so on. There are infinitely many of them. If is an integer, however large, we can always add 1 to to get , which is even larger.
If and are two integers, we write divides . This means is an integer multiple of . For instance, it is true that but not true that .
Prime Numbers
Of course, number 1 is a trivial (non-interesting) divisor of every integer. It is also true that for every integer , we have .
Prime numbers are characterized by the fact that they do not have any other divisors except the trivial ones: 1 and themselves.
There is an interesting fact that we will use below. Take two consecutive integers and greater than 1 divides both and . Then it would also divide the difference . But this difference is 1. No number greater than 1 divides 1. End of proof.
Euclid's Fundamental Theorem of Arithmetic
We also need Euclid's Fundamental Theorem of Arithmetic [1, p. 26]:
Every positive integer can be written uniquely (up to order) as the product of prime numbers.
Here, one may wonder about the number 1. 1 certainly a positive integer. There are two ways to address this. One way is to restrict positive integers to those greater than 1, as we did in the beginning. The other way is more tricky. In mathematics, the product of numbers in the empty set is considered, by a convenient definition, to equal 1. So, 1 is the product of an empty set of primes!
Building the Prime Number Factory
It is time now to build our factory of prime numbers based on the old idea in Euclid's "Elements" (circa 300 BC). The fact that such a factory exists means that the number of primes is infinite, thus there is no such thing as the Last Prime Number. The factory's principle is this: for any finite collection of prime numbers, it constructs another prime number, which was not in the collection. We can add this number to get an even bigger collection and run our machine again. Ad infinitum. Here's how it works:
Euclid's Prime Numbers Producing Machine
Let be any (finite, non-empty) collection of prime numbers. Take their product . Add 1 to to get . Then and are consecutive integers. Therefore, they do not have common non-trivial divisors. Now, there are two possibilities: either is a prime number, or it can be divided by some prime number . If it is a prime number, then this prime number is certainly bigger than any of the numbers . So we add it to our collection and get a larger collection. If is not a prime number, it is divisible by a prime number then cannot be in our previous collection because and have no common non-trivial divisor. We add to our collection and get a larger collection. End of proof.
Remark: If P is greater than 2, which is always the case with only one exception: when our original collection is just {2}, then, of course, we can replace in the machine going from to by are also consecutive integers.
An Example of the Machine in Action
Let us see how the machine works on an example (cf. [2,3]). Let's start with the following pair of prime numbers {5, 11}.
Then . Take . It is not a prime number. It has prime divisors 2 and 3. We add them to our collection to get {2, 3, 5, 11}. We multiply them to get a new . We subtract 1 to get 329. Again, it is not a prime number, but it is a product of two prime numbers 7 and 47. We add them to our collection, which becomes now {2, 3, 5, 7, 11, 47}.
We multiply them to get a new , this time . Again, we subtract 1 to get . Again, it is not a prime number, but it is a product of two prime numbers 151 and 719. We add them to our collection to get {2, 3, 5, 7, 11, 47, 151, 719}.
These were just a few steps, but the machine can work till the end of time, producing at each step new prime numbers. There is no “Last Prime Number.”
Remark: If P+1 is not a prime number, we can add to the collection not just one of its prime divisors, but all of them that are different from each other.
Remark: The machine is of course not very effective as it requires factorization of larger and larger numbers. But it does its job, which is: of proving the infinity of primes.
Exercise: Start just with {2} instead of (2,3,5}. Use the algorithm above. Describe what happens.
The Mysteries of Primes
The infinitude of primes has its mysteries. Mathematicians are always on the lookout for patterns and regularities in prime numbers. Recently, researchers from City University of Hong Kong and North Carolina State University released a paper entitled “Periodic Table of Primes" [4]. They argue that primes can be predicted using a periodic table-like structure. The date of appearance of this paper (April 6, 2024) differs only by the prime number 5 from April 1, which brings to mind some prime associations.
Summary
Prime numbers, those elusive building blocks of mathematics, have captivated minds from ancient Egypt to modern academia. With Euclid's timeless machine, we've shown that the hunt for primes is a never-ending adventure. As mathematicians continue to unravel their mysteries, who knows what prime secrets await discovery? So, next time you ponder numbers, remember: there's always another prime just around the corner. Happy number crunching!
References
[1] Melvyn B. Nathanson, Elementary Methods in Number Theory, Springer, 2000.
[2] M. Hardy, C. Woodgold, Prime Simplicity. Math Intelligencer 31, 44–52 (2009). https://doi.org/10.1007/s00283-009-9064-8
[3] M Hardy, Notes 98.20 A concrete view of Euclid’s proof of the infinitude of primes, The Mathematical Gazette, volume 98, issue 543 (2014), https://doi.org/10.1017/s0025557200008172.
[4] Han-Lin Li, Shu-Cherng Fang, Way Kuo, The Periodic Table of Primes, Date Written: April 6, 2024, https://doi.org/10.2139/ssrn.4742238.
P.S. 04-0824 And so one of my blog posts, Reflections on the current model of cosmology - By Thomas Görnitz, has been translated into Russian and reposted by V.V. Varlamov on his blog:
P.S. 05-08-24 And so again: my blog is being quoted in Cyrillic:
Starting with {2} and always calculating P+1:
ReplyDeleteIteration1: prime_set={2}, P=2, P+1=3, 3 is prime, so add it to the set.
Iteration2: prime_set={2,3}, P=6, P+1=7, 7 is prime, so add it to the set.
Iteration3: prime_set={2,3,7}, P=42, P+1=43, 43 is prime, so add it to the set.
Iteration4: prime_set={2,3,7,43}, P=1806, P+1=1807, 1807=13*139, so add 13 and 139 to the set.
Iteration5: prime_set={2,3,7,13,43,139}, and so on.
Starting with {2} and always calculating P-1:
Iteration1: prime_set={2}, P=2, P-1=1, 1 is prime, so add it to the set.
Iteration2: prime_set={1,2}, P=2, P-1=1, 1 is prime, and is already in the set. Stop.
Good, Just one comment: according to the definition adopted in the post number 1 is not counted as prime.
DeleteMy first iteration was slightly different, and doesn't stop in the case of P-1.
DeleteIteration1: P={2}, P=(2*2)=4. P-1=3, therefore P={2,3} ...
I don't know if it is correct to do this, but it seems to me that the only way to make a product from a one-element set is to multiply it by itself. If this is correct, it may begin to explain why 1 is not included in the definition of primes, as P={1} does not 'start the factory' in the way P={2} does.
You take product of all elements in the set. If there is only one element, then the product of elements is just this one element. But you can replace {2} with {2,2} if you wish. Then your calculation is ok.
DeleteAh, in that case Natus' solution is correct. Thanks Ark!
DeleteRegarding the recent paper by researchers from City University of Hong Kong and North Carolina State University entitled “Periodic Table of Primes" [4]. Most everything in that paper is not new information, they just make it way more complicated than it needs to be. The authors are just basically working with Liu Fengsui's Prime Formula ( https://www.primepuzzles.net/problems/prob_037.htm ) from back in March of 2001. The primepuzzles post is a little rough but Liu later published a formal paper in a math journal in a year or two after that. Liu's paper contains the said author's CTC cyclic table of composites and PTP Periodic Table of Primes. It is all just LCM sets. All LCM sets are periodic at their primorial and symmetric about the 1/2 primorial. The authors' 48 roots is rather arbitrary as they could have chosen the {1,5} table of 2 roots at the 3 primorial (length of 6) or the {1,7,11,13,17,19,23,29} table of 8 roots at the 5 primorial (length of 30). They just went one step further to the 7 primorial (length of 210) with its 48 roots. They could keep going to any primorial length. The 11 primorial (length 2310) 480 roots, the 13 primorial (length 30,030) with its 5,760 roots, the 17 primorial (length 510,510) with its 92,160 roots. It is just another way of saying for example using the {1,5} table of 2 roots at the 3 primorial (length of 6), all primes greater than 3 must be of the form 6n+-{1,5}, which is 6n+-1. Or for the {1,7,11,13,17,19,23,29} table of 8 roots at the 5 primorial (length of 30), all primes greater than 5 must be of the form 30n+-{1,7,11,13,17,19,23,29}, and on and on. If you really want to go back, a case can be made that Eratosthenes demonstrated all of this with his sieve. If you examine the sets removed in the sieve they are just the LCM sets that are periodic and symmetric and the remainder set also becomes periodic and symmetric. The remainder set is just the said author's Periodic Table of Primes at each step of the way.
ReplyDeleteThank you very much for your enlightening comments about the Chinese periodic primes table paper.
DeleteDon Christmann: "Or for the {1,7,11,13,17,19,23,29} table of 8 roots at the 5 primorial (length of 30), all primes greater than 5 must be of the form 30n+-{1,7,11,13,17,19,23,29}, and so on."
DeleteWow. So this is what Gary Croft (from https://www.primesdemystified.com) used to claim that "the invisible hand of 2, 3 and 5, and their factorial 30, create the structure within which the balance of the prime numbers, i.e., all those greater than 5, are arrayed algorithmically–as we shall demonstrate. Primes 2, 3 and 5 play out in modulo 30-60-90 cycles (decomposing to {3,6,9} sequencing at the digital root level). Once the role of 2, 3 and 5 is properly understood, all else falls beautifully into place."
"The first rotation of the sieve, comprised of 8 members, (1, 7, 11, 13, 17, 19, 23 and 29), is the deterministic key to everything that follows. These are the first 8 counting numbers not divisible by 2, 3 or 5, a sequence which by definition includes (and only includes) 1 and all primes ≥ 7 and their multiplicative multiples (and, as you'll see below, it's conjectured that the entire set can be generated by a simple expression involving 2, 3 and 5). The inference of this conjecture is that all prime numbers greater than 5, i.e., starting with 7, can be produced by this expression."
"Croft first made his Prime Spiral Sieve 'public' in July, 1993, when he presented it to City University math instructor Mark Wahl while attending his class "Mathematics: Its Meaning, Mystery and Methods" (July/August, 1993)."
Yes, this is precisely why Croft's observations work and this new paper just went one step further out of an infinite number of steps and made their similar observation at the next step in the process. Once again I would say that Eratosthenes first exposed the information, not explicitly, but if one actually examined the Eratosthenes sieve and the removal sets and remainder sets at each step, the information is right there already. It is interesting in how sometimes there is much more information hidden in plain sight in the old masters' works, if we just look a little deeper. It is not that new information cannot arise from research like this new paper, but it is amazing that we can look back and see that is was already sitting there decades, centuries and even millennia ago.
DeleteSession 7 November 1998:
ReplyDelete"A: Mathematics converts to sound in geometric measurements."
What if musical ratios were only made of prime numbers?
Consider the following matrices:
[1/1]
unique_set = {1}
[1/1, 2/1
1/2, 2/2]
unique_set = {1/2,1,2}
[1/1, 2/1, 3/1
1/2, 2/2, 3/2
1/3, 2/3, 3/3]
unique_set = {1/3, 1/2, 2/3, 1, 3/2, 2, 3}
[1/1, 2/1, 3/1, 5/1
1/2, 2/2, 3/2, 5/2
1/3, 2/3, 3/3, 5/3
1/5, 2/5, 3/5, 5/5]
unique_set = {1/5, 1/3, 2/5, 1/2, 3/5, 2/3, 1, 5/3, 3/2, 2, 5/2, 3, 5}
[1/1, 2/1, 3/1, 5/1, 7/1
1/2, 2/2, 3/2, 5/2, 7/2
1/3, 2/3, 3/3, 5/3, 7/3
1/5, 2/5, 3/5, 5/5, 7/5
1/7, 2/7, 3/7, 5/7, 7/7]
unique_set = {1/7, 1/5, 2/7, 1/3, 2/5, 3/7, 1/2, 3/5, 2/3, 5/7, 1, 7/5, 5/3, 3/2, 2, 7/3, 5/2, 3, 7/2, 5, 7}
...and so on.
The number of unique ratios of each matrix corresponds to the sequence <1,3,7,13,21,...> which expresses the formula a(n) = n^2-n+1, starting from n=1. This formula corresponds to the Central Polygonal Numbers — a link to geometry!
"Central polygonal numbers are a class of figurate numbers, each formed by a central dot, surrounded by polygonal layers with a constant number of sides. Each side of a polygonal layer contains one more dot than a side in the previous layer, starting from the second polygonal layer. This unique structure gives rise to a distinct sequence of numbers."