1 The Periodic Table of Primes Han-Lin Li1, Shu-Cherng Fang2, Way Kuo3* 1Department of Computer Science, City University of Hong Kong 2Department of Industrial and Systems Engineering, North Carolina State University 3Hong Kong Institute for Advanced Study, City University of Hong Kong *Corresponding author: Way Kuo, way@cityu.edu.hk Abstract Prime numbers are roots of integers1. Over millennia, nobody has been able to predict where prime numbers sprout2 or how they spread. It has been believed that primes grow like weeds among natural numbers. This study nevertheless establishes the Periodic Table of Primes (PTP) using four prime numbers of 2, 3, 5, and 7. We identify 48 integers out of a period of 2 × 3 × 5 × 7 = 210 to be the roots of all primes and composites without factors of 2, 3, 5, and 7, each of which is an offspring of the 48 integers uniquely allocated on the PTP. The PTP provides a platform to make the study of primes clearer and easier. In other words, the 48 integers serve as the genes of primes, and among them 15 pairs serve as the genes of twin primes. This study identifies the cyclic effects among the composites, spawning primes distributed periodically. Three major contributions made in the article are the Formula of Primes, the Periodic Table of Primes, and the Counting Functions for Primes and Twin Primes. These findings provide answers to questions of interest to the academic community, such as finding a future prime, factoring an integer, illustrating the fundamental theorem of arithmetic, predicting the total number of primes and twin primes or estimating the maximum prime gap within an interval, among others. Introduction The fundamental theorem of arithmetic is one of the greatest theorems of mathematics1, which states that a composite integer can be represented uniquely as a product of primes. In recent years, primes as roots of integers have been studied and widely applied to data science1, cryptography2, and engineering design3, etc. For a long time, many said these integer roots grew like weeds among natural numbers, but nobody could predict where the next primes may sprout4. Many believe that primes are unpredictable. Oliver and Soundaranjan5 investigated the distribution of consecutive primes, while Luque and Lacasa6 reported patterns of the last digit of primes. Wang7 adopted pictures to search for regularities of some of the primes, but with significant bias. Owing to the lack of insight, there is no effective computable formula for prime listing, prime factorization, counting functions for primes and twin primes, or the next prime of a known number. Studies published to predict primality and pattern of primes are ad hoc with tremendous limitations and uncertainties. Some6 observed that primes located near each other tend to avoid repeating their last digits, which indicates that primes are not distributed as random as theorists often assume. Others such as Tóth8 found the existence of the primes in k-tuple where the difference of two neighboring primes stays the same. However, few are able to explain what and where these tuple primes occur, nor can they predict the longest tuple primes within an interval. In addition to the work by Zhang9, Maynard10 reported the minimum upper bound for prime gaps. Also reported includes applications of the Maynard-Tao sieve method11. Few spells out Electronic copy available at: https://ssrn.com/abstract=4742238 2 the locations where a maximum prime gap in an interval will appear. Furthermore, there is no study or intent on visualizing the fundamental theorem of arithmetic to illustrate the spatial relationship between a composite number and its factors. Initialized by Gauss, there were studies on counting the total numbers of primes and twin primes within an interval, but only approximations were obtained 12,13. Since primes are the roots of integers, can we identify the roots of primes? Can we find rules for generating composites; and if possible, may we say that, in some aspects, primes are predictable? Li et al.14 used 2, 3 and 5 in building a universal color system C235 to unify RGB (a light color frame) and CMYK (a pigment color frame). C235 represents colors R(red), G(green), and B(blue) by primes 2, 3, and 5, respectively. Consequently, C(cyan), M(magenta), Y(yellow), and K(gray) are represented by 3 × 5 = 15, 2 × 5 = 10, 2 × 3 = 6, and 2 × 3 × 5 = 30, respectively. Through this transformation, all colors are representable by 7 root numbers of 2, 3, 5, 6, 10, 15, and 30. These root numbers encode millions of colors on a color wheel 14. Inspired by C235, we intend to: • • • find a set of integers which serves as the root of primes and twin primes, establish the Formula of Primes and build the Periodic Table of Primes (PTP) that allocate primes, and form the Counting Functions for Primes and Twin Primes. Also accomplished include to predict within an interval, the largest k-tuple primes8 with the same difference as well as to find the conditions for a maximum prime gap to occur, and to visualize the fundamental theorem of arithmetic. First, we define a concise strategy by selecting the roots of primes greater than 10 and composites without factors of 2, 3, 5, and 7, and then form the Cyclic Table of Composites (CTC) identifying the locations of composites, followed by advancing the Formula of Primes and building the PTP. The procedures are summarized in 4 steps: 1. Selecting the roots and cycles Adopting the first four primes 2, 3, 5, and 7, we take 2 × 3 × 5 × 7 = 210 as the length of a period. Within the interval [11, 211], we sort out 48 integers of primes or composites that do not contain factors of 2, 3, 5, and 7. These 48 integers, to generate primes and composites, are considered as the roots denoted by r1 = 11, r2 = 13, r3 = 17, โฏ , r23 = 103, r24 = 107, โฏ , r48 = 211. These r๐ s are placed on the left column of a table. 2. Developing the Cyclic Table of Composites (CTC) The CTC consists of multiple 48 × 48 matrices [๐(i, j)] derived from the positions of composites without factors of 2, 3, 5, and 7. Within an interval of [48(θ − 1) + 1, 48θ], where θ ∈ N+ is considered a cycle, there are 48 integers which do not contain factors of 2, 3, 5, and 7. These 48 × θ integers are either primes or composites, denoted as q1 = r1 = 11, q2 = ๐2 = 13, q3 = r3 = 17, โฏ , q48 = r48 = 211, for cycle 1; q49 = q1 + 210 = 221, q50 = q2 + 210 = 223, โฏ , q96 = q48 + 210 = 421, for cycle 2; … ; q48(θ−1)+1 = q48(θ−2)+1 + 210 = q1 + 210(θ − 1), q48(θ−1)+2 = q48(θ−2)+2 + 210 = q2 + 210(θ − 1), โฏ , and q48θ = q48(θ−1) + 210 = q48 + 210(θ − 1), for cycle θ . We call q1 to q48 segments of cycle 1, q49 to q96 segments of cycle 2, …, and q48×(θ−1)+1 to q48θ segments of cycle θ. The qj s are placed on the top row of a CTC table of θ cycles. Formulating the CTC of the first cycle, namely, CTC(1), we observe the dual effect existing in [๐(i, j)], where index i refers to ri and j refers to qj , such that in each row of this Electronic copy available at: https://ssrn.com/abstract=4742238 3 table, pairs of entries share the same value of elements ๐(i, j)s. We also find for each row in the CTC(1), there is another row complementary to it, calling it the mirror effect. Specifically, there exist the mirror effects between the 23 top rows and the next 23 rows as well as between the 47th row and the 48th row of CTC(1). We define the basic CTC (CTC(basic)) to be a 24 × 48 matrix of [๐(i, j)], i = 1, 2, โฏ , 23, 47 and j = 1, 2, โฏ , 48. Utilizing the mirror effect, we find the complementary CTC (CTC(complementary)) a 24 × 48 matrix made of [๐(i, j)] for i = 24, 25, โฏ , 46, 48 and j = 1, 2, โฏ , 48. Combining the CTC(basic) and the CTC(complementary), we form CTC(1). We observe further an intermedia effect in CTC(1) and a cyclic effect between different cycles of composite tables where each cycle is a 48 × 48 matrix. The entries of a higher cycle are self-reproductive of CTC(1). By utilizing the intermedia and the cyclic effects, we generate the CTC with multiple cycles, CTC(1), CTC(2), CTC(3), โฏ, CTC(θ). 3. Advancing the Formula of Primes By deleting all composites located by CTC in an interval, we develop the Formula of Primes. Given a positive integer b with b – 211 being a multiplier of 210, an integer α ∈ [0, b] without factors of 2, 3, 5, and 7 is a prime if and only if there exists an i ∈ {1, 2, … , 48} such that α = ri + 210k, for some integer k with k + 1 ∉ L๐ (i), where L๐ (i) is a set of ๐(i, j)s which satisfy some conditions of b and θ described in the Results section. 4. Building the Periodic Table of primes (PTP) Denote PTP(k 0 , k′) as a periodic table of primes from periods k 0 to k′, k 0 < k′ ∈ N+ . PTP(k 0 , k′) is a 48 × (k ′ − k 0 + 1) matrix composed of all primes within an interval [a, b], for a = 11 + 210 × k 0 and b = 211 + 210 × k′. Denote PTP + (k 0 , k′) as a matrix composed of primes in PTP(k 0 , k′) plus composites without factors 2, 3, 5, and 7 within the interval [a, b]. The steps of building a PTP are depicted in Figure 1. Starting from the CTC(initial), we form a CTC(initial dual). By combing the CTC(initial) and the CTC(initial dual), we obtain the CTC(basic) which is outlined in the Supplement under Establishing the CTC(basic) by the CTC(initial). Once CTC(1) is deduced from CTC(basic), all subsequent CTC(2), CTC(3), …, CTC(θ) are readily available. Utilizing CTCs, we generate PTP+(k 0 , k′), where we delete all composites to obtain PTP(k 0 , k′). The red and green points in Figure 1 represent the diagonal and intermediate effects, respectively. The red and green points also show the cyclical pattern on CTC and the periodical pattern on PTP+. Based on the CTCs and the PTPs, we form the Counting Functions to predict the total number of primes and twin-prime pairs within an interval, instead of just an approximations13. Methods and Observations Without loss of generality, discussions in this section on the CTC, the CTC(basic) and the CTC(complementary) are referred to cycle 1 unless stated otherwise. In CTC, consisting of 48 rows and 48 columns, denote r๐ as the ith row and qj the jth column. Let Electronic copy available at: https://ssrn.com/abstract=4742238 4 S1 = = S2 = = {r1 , r2 , … , r23 , r47 } {11, 13, 17, 19, 23, 29, 31, 37, 41, 43, 47, 53, 59, 61, 67, 71, 73, 79, 83, 89, 97, 101, 103, 209}, {r24 , r25 , … , r46 , r48 } {107, 109, 113, 121, 127, 131, 137, 139, 143, 149, 151, 157, 163, 167, 169, 173, 179, 181, 187, 191, 193, 197, 199, 211}, and S = S1 ∪ S2 = {r1 , r2 , … , r48 } ={q1 , q2 , … , q48 }. Note that r๐ s and qj s are integers in [11, 211] without factors of 2, 3, 5, and 7. Let (i, j) be at a position in the CTC and ๐(i, j) be the corresponding entry of point (i, j), where 1 ≤ i, j ≤ 48, determined by ๐(i, j) = 1 + qj ×qh(i,j) −r๐ 210 , โฏ Expression (1) where qh(i,j) ∈ S is a minimum integer such that qj × q๐ก(๐ข,๐ฃ) − r๐ is a multiplier of 210. Expression (1) states that a composite β, without factors of 2, 3, 5 and 7, generated by β = qj × q๐ก(๐ข,๐ฃ) , q๐ก(๐ข,๐ฃ) being one of {q1 , q2 , … , q48 }, is the ((๐(i, j) − 1)th descent of ri. If ๐(i, j) = 1, then β = r๐ and β is the root ri itself. In general, if ๐(i, j) = n, then β = r๐ + 210 × (n − 1) is the nth descendant of root r๐ . All CTC(θ)s are similarly specified. An outline of CTC is shown in Figure 2. Various effects in CTC are described by the following statements. Presented in Table S1(a) along with an illustrative example are given in the Supplement under Example for Table S1(a). Observing Table S1(a), note that ๐(i, i) = qi + 1, ๐ = 1, 2, … , 23, 47. We identify the diagonal effect in CTC in Statement 1 below. Statement 1 (The Diagonal effect of the CTC) In the CTC, ๐(๐ฃ, ๐ฃ) = ๐ช๐ฃ + ๐, for ๐ฃ = ๐, ๐, โฏ , ๐๐. Moreover, ๐(๐ฃ, ๐ฃ) ≥ ๐(๐ข, ๐ฃ), for ๐ข, ๐ฃ = ๐, ๐, โฏ , ๐๐ ๐๐ง๐ ๐ข ≠ ๐ฃ. Analyzing the distribution of ๐(i, j) in the ith row of Table S1(a), we find the dual effect between ๐(i, j) and ๐(i, h), described in Statement 2 below. Statement 2 (The Dual effect of the CTC) In the CTC, for any ๐(๐ข, ๐ฃ), ๐ข, ๐ฃ = ๐, ๐, โฏ , ๐๐ ๐๐ง๐ ๐ข ≠ ๐ฃ, there exists an ๐ก(๐ข, ๐ฃ) ∈ {๐, ๐, โฏ , ๐๐} such that ๐(๐ข, ๐ฃ) = ๐(๐ข, ๐ก(๐ข, ๐ฃ)). Statement 2 implies that given ๐(i, j), we know ๐(I, h(i, j)), under which situation j and h(i, j) are dual to each other. Take ๐ = 1 and r1 = 11 for instance, ๐(1, 5) = ๐(1, 8) = 5, where q5 × q8 = 23 × 37 = 851 = 11 + 210(5 − 1). Notice that when h(i, j) = j, qj 2 is a descendant of r๐ and ๐(i, j) appears in the ith row only once. For instance, ๐(48,9) = 8 appears only once and 412 = 1682 = 211 + 210× (8 − 1). Further elaboration of the dual effect and associated examples are given in the Supplement under Elaboration of the Dual Effect of the CTC(basic). Analyzing Tables S1(a) and S1(b), we also identify the mirror effect between each row of the CTC(basic) and its complementary row in CTC deduced below. Electronic copy available at: https://ssrn.com/abstract=4742238 5 Statement 3 (The Mirror effect in the CTC) The mirror effect in the CTC exists between pairs of all rows such that (i) ๐(๐๐ − ๐ข, ๐ฃ) + ๐(๐ข, ๐ฃ) = { (ii) ๐(๐๐, ๐ฃ) + ๐(๐๐, ๐ฃ) = { ๐ช๐ฃ + ๐, ๐ข๐ ๐ข ≠ ๐ฃ for ๐ข = ๐, ๐, โฏ ๐๐, ๐ฃ = ๐, โฏ , ๐๐. ๐๐ช๐ฃ + ๐, ๐ข๐ ๐ข = ๐ฃ, ๐ช๐ฃ , ๐ข๐ ๐ฃ = ๐, โฏ , ๐๐ ๐๐ช๐ฃ , ๐ข๐ ๐ฃ = ๐๐, ๐๐ Statement 3 implies that if we know ๐(i, j) for 1 ≤ i ≤ 23, then we can find ๐(47 − i, j). Similarly, if we know ๐(47, j), then we find ๐(48, j). Following Statement 3, we form the CTC(complementary), which is the complementary of the CTC(basic). CTC(complementary) is presented as Table S1(b), in the Supplement. Merging the CTC(basic) and the CTC(complementary), we obtain the CTC(1) as the CTC for cycle 1, referring to Figure 1. Consider CTC in Figure 2. There are two red diagonals shown in Figure 2, which further elaborates Table S1(a) and Table S1(b), The first diagonal is ๐(1, 1) = 12, ๐(2, 2) = 14, โฏ , and ๐(46, 46) = 200, and the second one is ๐(46, 1) = 11, ๐(45, 2) = 13, โฏ, and ๐(1, 46) = 199. The two cross lines in Figure 2 are obvious due to the mirror effect. ๐(47, j) and ๐(48, j), j = 1, 2, 3, โฏ , 48, are mirror with each other. An Example is given in the Supplement under An Associated Example for Statement 3. Table S1(a) and Table S1(b) show another distribution of ๐(i, j) on CTC, named the intermedia effect, which happens at some symmetrical columns, described in the Supplement under Statement and Proof on the Intermedia effect in the CTC. Both the diagonal effect and the intermedia effect on the CTC are helpful to show how to transfer CTC to PTP. In Figure 2, the red and green numbers represent the diagonal and the intermediate effects, respectively. Expanding Expression (1) by denoting ๐(i, j + 48(θ − 1)) as the entries of CTC(θ), where θ ∈ N+ , we find the cyclic effect between CTC(1) and CTC(θ) made in Statement 4. Part of the upper and lower rows of CTC(2) are illustrated in Table S2(a) and Table S2(b), respectively. Statement 4 (The Cyclic effect between CTC(1) and CTC(๐)) The cyclic effect exists between CTC(1) and CTC(๐), ๐ = 2, 3, …, such that ๐(๐ข, ๐ฃ + ๐๐(๐ − ๐)) = ๐(๐ข, ๐ฃ) + (๐ − ๐)๐ช๐ก(๐ข,๐ฃ) , ๐๐จ๐ซ ๐ ≤ ๐ข, ๐ฃ ≤ ๐๐, where ๐(๐ข, ๐ฃ) = ๐ + ๐ช๐ฃ ×๐ช๐ก(๐ข,๐ฃ) −๐ซ๐ ๐๐๐ is the (i, j) entry in CTC(1). Through ๐(i, j), ri , qj at cycle 1 (i.e., θ = 1), we find ๐(i, j + 48(θ − 1)) for cycle θ. An example is given in the Supplement under Numerical Illustrations for Statement 4. Proofs for the four above Statements are given in the Supplement under Proofs of the Statements. With the dual, the diagonal, the mirror, the intermedia, and the cyclic effects based on the 48 roots, composites without factors of 2, 3, 5, and 7 are distributed periodically and cyclically. Results This study provides three major results: the Formula of Primes, the Periodic Table of Primes, and the Counting Functions for Primes and Twin Primes. Electronic copy available at: https://ssrn.com/abstract=4742238 6 Given a positive integer b with b – 211 being a multiplier ๏ฑ๏ช of 210, and i ∈ {1, 2, โฏ , 48}, define L๐ (i) ≡ {๐(i, j + 48(θ − 1)|j = 1, 2, โฏ , 48, and θ = 1, 2, … , ๏ฑ๏ช ๏ฝ. (2) โฏExpression Specifically, L๐ (i) is a set consisting of ๐(i, j) ≤ โi,θ+1 , for j = 1, 2, …, 48θ, where โ๐,θ = min{๐(i, j)|j = 48(θ − 1) + 1, 48(θ − 1) + 2, โฏ , 48θ, for θ ∈ N+ }, and 211โi,θ+1 ≥ b . j Notice that ๐(i, j + 48(θ − 1)) = ๐(i, j) + (θ − 1)q๐ก(๐ข,๐ฃ) where q๐ก(๐ข,๐ฃ) is derived from Expression (1). For any positive integer α containing no factors of 2, 3, 5, and 7, α must have a unique root r๐ ∈ S such that α − r๐ is a multiple of 210. Moreover, if α is not a composite number, then it is a prime. From Expression (2) and Statement 4, we know that for any composite β ≤ b, which does not contain factors of 2, 3, 5, and 7, it can be expressed by β = qj × q๐ก(๐ข,๐ฃ) = ri + 210[๐(i, j) + (θ − 1)q๐ก(๐ข,๐ฃ) − 1] = ri + 210[๐(i, h(i, j)) + (θ − 1)qj − 1], where ๐(i, j) + (θ − 1)q๐ก(i,๐ฃ) and ๐(i, h(i, j)) + (θ − 1)qj ∈ Lb (i). We summarize the Formula of Primes below: The Formula of Primes An integer ๐ ∈ [๐๐, ๐] containing no factors of 2, 3, 5, and 7 is a prime if and only if there exists ๐ซ๐ข ∈ ๐ and ๐ ∈ N0 ๐ฌ๐ฎ๐๐ก that ๐ = ๐ซ๐ข + ๐๐๐๐ค, with k + 1 ∉ ๐๐ (๐ข), ๐ข = ๐, ๐, … , ๐๐. An integer α = ri + 210k ≤ b is a prime, if there does not exist a non-negative integer m which satisfies k + 1 = ๐(i, j) + m × q๐ก(๐ข,๐ฃ) , for i = 1, 2, …, 48. A list of L๐ (1) for various b up to 44521 is listed in Table S3, which is presented in the Supplement under Justification for Establishing Lb (1) and Table S3. According to CTC (θ) and the Formula of Primes, we specify the Periodic Table of Primes: The Periodic Table of Primes from period ๐ค ๐ ๐ญ๐จ ๐ฉ๐๐ซ๐ข๐จ๐ ๐ค ′ For given ๐ค ๐จ ∈ ๐๐ , ๐ค ′ ∈ ๐+ , ๐๐ง๐ ๐ค ′ > ๐ค ๐จ , ๐TP (๐ค ๐ , ๐ค ′ ) is a ๐๐ × (๐ค ′ − ๐ค ๐ + ๐) matrix [๐(๐ข, ๐ค + ๐๐๐)], where ๐(โ) is an integer such that ๐๐ + ๐๐๐ × ๐ค ๐ ≤ ๐(๐ข, ๐ค + ๐๐๐) = ๐ซ๐ + ๐๐๐๐ค ≤ ๐๐๐(๐ค ′ + ๐), for k ∈ ๐๐ ๐ฐ๐ข๐ญ๐ก ๐ค + ๐ ∉ ๐๐ (๐ข) ๐๐ง๐ ๐ = ๐๐๐ (๐ค ′ + ๐). We denote PTP + (k o , k ′ ) for a matrix composed of primes in PTP(k o , k ′ ) plus composites without factors of 2, 3, 5 and 7, within an interval [11 + 211 × k o , 211(k ′ + 1)]. Figure 3 is an outline of PTP+(0, 210), which includes PTP(0, 210) and composites without factors of 2, 3, 5, and 7 within the interval [11, 1 + 2102 ] converted from Figure 2. Figure 3 has 48 rows and 211 columns, for i = 1, 2, โฏ , 48, and k = 0, 1, 2, 3, โฏ , 210. λ(i, k) is an integer of the entry (i, k), computed from ๐(i, j) of Figure 2. The two dotted red curves in Figure 3 are configured from two diagonal red lines of Figures 2, i.e., 12–14–18 โฏ 198–200 and 11–13–17 โฏ 197–199. The two dotted green diamond-shape lines in Figure 3 come from two green lines of Figure 2 of 24–75–74 โฏ 29–80 and 84–31–32 โฏ 77–24. Electronic copy available at: https://ssrn.com/abstract=4742238 7 Table 1 is a realization of PTP+(0,10) including primes and composites without factors of 2, 3, 5, and 7. Further elaboration is seen in the Supplement under Numerical Illustrations for The Periodic Table of Primes. According to Gauss12,13, the number of primes no more than b is approximately b ln b . We find further the exact Counting Function for Primes π(b) when b - 211 is a multiplier of 210: The Counting Function for Primes ๐๐ ๐(๐) = ๐ + ๐๐ × (๐ค + ๐) − ∑|๐๐ (๐ข)|. โฏ ๐๐ฑ๐ฉ๐ซ๐๐ฌ๐ฌ๐ข๐จ๐ง (๐) ๐=๐ Currently, few predict the number of twin-prime pairs in an interval [2, b]. For b – 211 as a multiplier of 210, denote T(i, i + 1) as the number of twin-prime pairs on PTP(0, ⌈ b 211 ⌉), i.e., T(i, i + 1) is a set of k with k + 1 ∉ Lb (i) and k + 1 ∉ Lb (i + 1), where i and i + 1 are for the ith and the (i + 1)st row on the PTP, respectively. Our study finds the exact Counting Function for Twin Primes π∗ (b) below: The Counting Function for Twin Primes ๐∗ (๐) = ๐ + ∑|๐(๐ข, ๐ข + ๐)|, ๐ข∈๐ โฏ ๐๐ฑ๐ฉ๐ซ๐๐ฌ๐ฌ๐ข๐จ๐ง (๐) where ๐ = {๐, ๐, ๐, ๐, ๐๐, ๐๐, ๐๐, ๐๐, ๐๐, ๐๐, ๐๐, ๐๐, ๐๐, ๐๐, ๐๐}. Proofs of Expressions (3) and (4) are given in the Supplement under Proofs of Expressions (3) and (4), and examples of computing π(b) and π∗ (b) are illustrated in the Supplement under Nine Examples for the Predictions of Primes by the PTP. Discussion The unpredictability of prime numbers forms a basis of many applications, one being encryption called the RSA algorithm 2. However, neither the ancient Sieve of Eratosthenes’ 15 nor the modern Sieve of Atkin’s15 algorithms have ever elaborated the physical meaning of finding primes. Torquato, Zhang and Courcy-Ireland16 claimed that they found a physical structure pattern hidden in the distribution of prime numbers, but that discovery still didn’t explain the essence of prime numbers. This paper identifies 48 natural numbers between 11 and 211, which do not contain factors of 2, 3, 5, and 7, to be the roots for generating all primes and composites without factors of 2, 3, 5, and 7. The locations of such composites exhibit periodic and cyclic properties, as represented by the CTC, which enable us to eliminate them for finding primes, as represented by the PTP. Treating the 48 roots as the genes of prime numbers, we can easily find the next prime of any given prime number and identify the next pair of twin primes. Our findings provide a platform to study many primes-related problems. No primes, twin primes or primesrelated issues can ever surface if such issues are not rooted to the 48 integers. After all, prime numbers are not as random as many believe. We form the CTC, followed by the PTP. All these present the primes effectively and with physical meaning. In addition, we can count the exact numbers of primes and twin primes within an interval. Concluded below are some further thoughts. 1. Instead of choosing 2, 3, 5 and 7, one may choose 2, 3, 5, 7, and 11 as the base. By so doing, the PTP will become gigantic, too complicated, and difficult to visualize, although Electronic copy available at: https://ssrn.com/abstract=4742238 8 likely more effective. If one is interested in the behaviors of super-large primes, then such a large prime table could be useful. 2. The PTP is helpful in explaining many unclear phenomena. For instance, it explains a troublesome observation6 from the past which for a given prime with the last digit of 1, the chance of its next prime to have 1 as its last digit is much less than 3 or 7 or 9. From the Formula of Primes, if a given prime is 221 (i. e. , θ = 2, r๐ = 11), then the most possible near primes should be firstly 223 (i. e. , θ = 2, r๐ = 13), followed by 227, 229, 233, 239, and 241. 3. From the 48 roots identified, we find no triplet or higher multiples of primes existing in the roots. Therefore, there will be no triplet primes found in future generations. Likewise, all twin primes will appear exactly at the parallel locations as those appearing in the 48 roots of the PTP. This implies that all twin-prime pairs are descendent of 15 pairs of twin primes or composites. In fact, an equal chance is found in the last digit of the 48 roots for 1, 3, 7, and 9. Therefore, the last digits for all primes will each have a 25% chance to be 1, 3, 7, and 9 when primes go to infinity. 4. After the work of Zhang9, Maynard10 claimed that most existing resources used in finding the infinitely achievable small gaps between primes adopted his approach. Based on the CTC and the PTP, we readily find the maximum prime gap for various intervals. This argument is amended under Justification of Claim 4 in Discussion as well as depicted in Figure S1(a) and Figure S1(b) of the Supplement. 5. Examples for the predictions of primes are given in the Supplement under Nine Examples for the Predictions of Primes by the PTP. Figures S1 and S2 in the Supplement are depicted for visualizing the fundamental theorem of arithmetic and identifying the maximum prime gap or the largest twin-prime pair in a specific interval, respectively. References 1. Cartesian orthogonal coordinate system. Encyclopedia of Mathematics. https://encyclopediaofmath.org/wiki/Cartesian_orthogonal_coordinate_system (2014). 2. Dickon, L. E., History of the Theory of Numbers, Volume II: Diophantine Analysis. Dover Publications, Inc. (2005). 3. Li, H. L., Huang, Y. H., Fang, S. C. & Kuo, W., A prime-logarithmic method for optimal reliability design. IEEE Trans. Reliability, 70, 146-162 (2021). 4. Zagier, D. The first 50 million prime numbers. The Mathematical Intelligencer 1 (Suppl 1), 7–19 (1977). https://doi.org/10.1007/BF03039306 https://people.mpim-bonn.mpg.de/zagier/files/doi/10.1007/BF03039306/fulltext.pdf 5. Oliver, R. J. L., & Soundaranjan, K., Unexpected biases in the distribution of consecutive primes, Proc of the National Academy of Sciences, 113, E4446E4454. (2016). https://doi.org/10.1073/pnas.1605366113 6. Luque, B. & Lacasa, L., The first-digit frequencies of prime numbers and Riemann Zeta zeros. Proceedings of the Royal Society A, 564, 2197-2216 (2009). https://doi.org/10.1098/rspa.2009.0126 7. Wang, X. The genesis of prime numbers—revealing the underlying periodicity of prime numbers. Advances in Pure Mathematics, 11, 12-18 (2021). DOI: 10.4236/apm.2021.111002. 8. Tóth, L., “On the asymptotic density of prime k-tuples and a conjecture of Hardy and Littlewood” Computational Methods in Science and Technology, 25(3) (2019). 9. Zhang, Y., "Bounded gaps between primes". Ann. of Math. 179 (3): 1121–1174 (2014). 10. Maynard, J. Small gaps between primes. Ann. Math. 181(2), 383–413 (2015). 11. Freiberg T., A note on the Theorem of Maynard and Tao, in Advances in the Theory of Numbers, Springer, (2015), https://link.springer.com/chapter/10.1007/978-1-49393201-6_4 Electronic copy available at: https://ssrn.com/abstract=4742238 9 12. Zagier, D., Newman’s short proof of the prime number theorem, American Math Monthly, 104(8), 705-708 (1997).JSTOR 2975232. 13. Nicomachus of Gerasa, Introduction to Arithmetic, 2 nd century. 14. Li, H. L., Fang, S. C., Lin, B. M. T. & Kuo, W., Unifying colors by primes. Light Sci Appl. 12, 32 (2023). https://doi.org/10.1038/s41377-023-01073-x 15. Atkin, A. O. L. & Bernstein, D. J., Prime sieves using binary quadratic forms, Math. Comp. 73, 1023-1030 (2004). 16. Torquato, S., Zhang, G. & Courcy-Ireland, M. de, Uncovering multiscale order in the prime numbers via scattering, J. of Statistical Mechanics: Theory and Experiment (2018). Electronic copy available at: https://ssrn.com/abstract=4742238 CTC(initial) ๐(๐ r๐ , ๐) j 1 2 q1 q2 โฏ 23 q23 1 r1 2 r2 โฎ seg root i dual effect r๐ j 24 25 q24 q25 โฎ โฎ โฎ CTC(complementary) j 1 2 me ๐(๐ nt qj ,๐ q1 q2 ) root i ri seg 1 r1 2 r2 โฏ 23 24 q23 q24 โฏ 46 47 48 q46 q47 q48 seg mirror effect me j nt ๐(๐ qj ,๐ root i ri ) 24 r24 25 r25 โฎ โฎ 1 2 โฏ 23 24 q1 q2 q23 q24 โฏ 46 47 48 q46 q47 q48 โฎ 23 r23 46 r46 47 r47 48 r48 CTC(1) CTC(2) seg 1 2 โฏ 23 24 โฏ 46 47 48 49 50 โฏ 71 72 โฏ 94 95 96 me nt q ๐(๐ q23 q24 โฏ q46 q47 q48 q49 q50 q71 q72 โฏ q94 q95 q96 root i r , ๐) q1 q2 1 r1 2 r2 CTC(θ) 48(θ-1)+1 q48(θ-1)+1 j ๐ โฎ โฎ 23 r23 24 r24 โฎ 10 46 q46 23 r23 CTC(basic) ๐ โฏ 1 r1 2 r2 23 r23 โฎ me j nt q ) root i me j nt q ,๐ ๐(๐ seg CTC(initial dual) 48θ-2 48θ-1 48θ q48θ-2 q48θ-1 q48θ cyclic effect โฎ 47 r47 48 r48 PTP+(0, 210) period k 1 2 ๐ (๐,k ) root i r๐ 1 r1 2 r2 โฏ PTP+(211, 420) 210 211 212 โฏ 420 PTP+(210(θ-1)+1, 210θ) โฏ 210(θ-1)+1 โฎ โฎ 23 r23 24 r24 โฎ โฎ 47 r47 48 r48 Figure 1: Schematic process of forming the PTP Note: for diagonal effect. for intermediate effect. Electronic copy available at: https://ssrn.com/abstract=4742238 θ≥2 210θ seg ๐ me nt ๐(๐ q๐ , root i r๐ ๐) Electronic copy available at: https://ssrn.com/abstract=4742238 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 11 13 17 19 23 29 31 37 41 43 47 53 59 61 67 71 73 79 83 89 97 101103107109113121127131137139143149151157163167169173179181187191193197199 209 211 1 11 12 2 2 9 5 16 7 5 7 29 24 33 40 39 37 52 59 57 63 9 67 88 24 84 16 37 111 33 39 29 40 67 57 52 63 84 59 145155145 88 155111182182199 199 12 2 13 10 14 13 7 12 24 25 36 23 38 41 34 31 57 7 29 43 60 29 59 79 12 75 31 98 23 43 85 34 59 85 131 13 75 60 25 24 79 10 41 38 98 131 57 197 36 197 14 3 17 18 74 32 193 193 18 4 19 20 22 86 191 191 20 5 23 24 21 87 187 187 24 6 29 30 10 71 35 100 181 181 30 7 31 32 19 89 179 179 32 8 37 38 9 69 37 101 173 173 38 9 41 42 68 38 169 169 42 10 43 44 16 92 167 167 44 163 163 48 11 47 48 15 93 12 53 54 7 65 41 103 157 157 54 13 59 60 12 96 151 151 60 14 61 62 6 63 43 104 149 149 62 15 67 68 10 98 143 143 68 16 71 72 9 99 139 139 72 17 73 74 60 46 137 137 74 18 79 80 7 101 131 131 80 CTC(initial) CTC(initial dual) 19 83 84 6 102 127 127 84 20 89 90 56 50 121 121 90 21 97 98 54 52 113 113 98 22 101 102 53 53 109 109 102 23 103 104107 107 104 24 107 103108 103 108 25 109 101 51 55 110 101 110 26 113 97 50 56 114 97 114 27 121 89 48 58 122 89 122 83 98 6 128 83 128 28 127 29 131 79 97 7 132 79 132 30 137 73 44 62 138 73 138 31 139 71 95 9 140 71 140 32 143 67 94 10 144 67 144 33 149 61 96 41 65 6 150 61 150 34 151 59 92 12 152 59 152 35 157 53 95 39 67 7 158 53 158 36 163 47 89 15 164 47 164 37 167 43 88 16 168 43 168 38 169 41 36 70 170 41 170 39 173 37 93 35 71 9 174 37 174 40 179 31 85 19 180 31 180 41 181 29 92 33 73 10 182 29 182 42 187 23 83 21 188 23 188 43 191 19 82 22 192 19 192 44 193 17 30 76 194 17 194 45 197 13 90 29 77 12 198 13 198 46 199 11 80 24 200 11 200 47 209 1 7 3 1 8 25 22 3 33 17 15 26 34 52 15 47 8 38 17 64 91 38 26 80 68 7 34 101 68 122 47 111 22 64 80 111101 33 159 52 25 122181159 91 181 210 210 48 211 10 6 14 18 15 4 9 34 8 26 32 27 25 9 52 24 65 41 66 25 6 63 77 27 41 106 87 26 63 15 92 32 127 87 77 52 66 136 14 127156 65 10 34 106 18 208 212 Note: Numbers in red — for diagonal effect. Numbers in green — for intermediate effect. Figure 2: An outline of CTC(1) 11 12 Electronic copy available at: https://ssrn.com/abstract=4742238 ๐ period k 1 2 3 โฎ ๐(๐, r๐ k) 23 48 72 96 110 120 11 13 17 21 22 23 24 25 26 โฎ 97 101 103 107 109 113 44 45 46 47 48 193 197 199 209 211 Note: 0 for diagonal effect. for intermediate effect. Figure 3: An outline of PTP+(0, 210) 130 140 150 160 170 180 190 210 Table 1: A Partial table of PTP+[0, 48) period k es/co mpo sites r๐ prim root i Electronic copy available at: https://ssrn.com/abstract=4742238 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 11 13 17 19 23 29 31 37 41 43 47 53 59 61 67 71 73 79 83 89 97 101 103 107 109 113 11x11 127 131 137 139 11x13 149 151 157 163 167 13x13 173 179 181 11x17 191 193 197 199 11x19 211 0 1 2 3 4 Note: Numbers in red — for composites 13 5 6 7 8 9 10 11 431 641 23x37 1061 31x41 1481 19x89 1901 2111 11x211 223 433 643 853 1063 19x67 1483 1693 11x173 2113 23x101 227 19x23 647 857 11x97 1277 1487 1697 1907 29x73 229 439 11x59 859 1069 1279 1489 1699 23x83 2129 233 443 653 863 29x37 1283 1493 13x131 1913 11x193 239 449 659 11x79 13x83 1289 1499 1709 19x101 2129 241 11x41 661 13x67 23x47 1291 19x79 29x59 17x113 2131 13x19 457 23x29 877 1087 1297 11x137 17x101 41x47 2137 251 461 11x61 881 1091 1301 1511 1721 1931 2141 11x23 463 673 883 1093 1303 1523 1723 1933 2143 257 467 677 887 1097 1307 37x41 11x157 13x149 19x113 263 11x43 683 19x47 1103 13x101 1523 1733 26x67 2153 269 479 13x53 29x31 1109 1319 11x139 37x47 1949 17x127 23x103 271 13x37 691 17x53 11x101 1321 1531 1741 1951 2161 277 487 17x41 907 1117 1327 29x53 1747 19x103 11x197 281 491 701 911 19x59 11x121 23x67 17x103 37x53 13x167 283 17x29 19x37 11x83 1123 31x43 1543 1753 13x151 41x53 17x17 499 709 919 1129 13x103 1549 1759 11x179 2179 293 503 23x31 13x71 11x103 17x79 1553 41x43 1973 37x59 13x23 509 719 929 17x67 19x71 1559 29x61 1979 11x199 307 23x37 727 937 31x37 23x59 1567 1777 1987 13x169 311 521 17x43 941 1151 1361 1571 13x137 11x181 31x71 313 523 733 23x41 1153 29x47 11x143 1783 1993 2203 317 17x31 11x67 947 13x89 1367 19x83 1787 1997 2207 11x29 23x23 739 13x73 19x61 37x37 1579 1789 1999 47x47 17x19 13x41 743 953 1163 1373 1583 11x163 2003 2213 331 541 751 31x31 1171 1381 37x43 1801 2011 2221 337 547 757 967 11x107 19x73 1597 13x139 2017 17x131 11x31 19x29 761 971 1181 13x107 1601 1811 23x47 23x97 347 557 13x59 977 1187 11x67 1607 23x79 2027 2237 349 13x43 769 11x89 29x41 1399 1609 17x107 2029 2239 353 563 773 983 1193 23x61 1613 1823 19x107 2243 359 569 19x41 23x43 11x109 1409 1619 31x59 2039 13x173 19x19 571 11x71 991 1201 17x83 1621 1831 13x157 2251 23x107 367 577 787 997 17x71 13x109 1627 11x167 23x89 37x61 373 11x53 13x61 17x59 1213 1423 23x71 19x97 2053 31x73 13x29 587 797 19x53 1217 1427 1637 1847 11x187 2267 13x13 379 19x31 17x47 1009 23x53 1429 11x149 43x43 29x71 2269 173 383 593 11x73 1013 1223 1433 31x53 17x109 2063 2273 179 389 599 809 1019 1229 1439 17x97 11x169 2069 43x53 181 17x23 601 811 1021 1231 11x131 13x127 1861 19x109 2281 11x17 397 607 19x43 13x79 1237 1447 1657 1867 31x67 2287 191 401 13x47 821 1031 17x73 1451 11x151 1871 2081 29x79 193 13x31 613 823 1033 11x113 1453 1663 1873 2083 2293 197 11x37 617 827 17x61 29x43 31x47 1667 1877 2087 2297 23x109 199 409 619 829 1039 1249 1459 1669 1879 2089 19x121 11x209 11x19 419 17x37 839 1049 1259 13x113 23x73 1889 2099 2309 211 421 631 29x29 1051 13x97 1471 41x41 31x61 11x191 2311 11 13 17 19 23 29 31 37 41 43 47 53 59 61 67 71 73 79 83 89 97 101 103 107 109 113 121 127 131 137 139 143 149 151 157 163 167 12 13 14 15 16 17 18 19 20 21 13x17 22 23 48 47x103 41x241 9883 9887 13x211 17x211 29x211 43x103 11x29x31 23x211 41x103 13x761 19x521 9901 9907 37x103 53x187 31x103 23x431 29x103 47x211 9923 9929 9931 19x523 9941 61x163 9949 37x269 23x433 9967 13x13x59 9973 11x907 17x587 67x149 97x103 13x769 73x137 10007 10009 17x19x31 43x233 11x911 37x271 29x107 79x127 10037 10039 31x107 37x107 11x11x83 13x773 41x107 19x23x23 23x209 19x209 43x107 17x109 13x209 47x107 89x113 10061 29x347 10067 10069 10079 17x593 14 Supporting Information for The Periodic Table of Primes 1. Establishing the CTC(basic) by the CTC(initial) Specifically, we adopt the following steps to establish the CTC(basic): a. From the CTC(basic) of Figure 1, we find 23 × 23 CTC(initial) matrix for the first cycle. In the CTC(initial), its entries of [๐(i, j)], for i = 1, 2, … , 23 and j = 1, 2, … , 23, are both starting from 11; namely, the left column contains r1, r2, …, r23 and the top row contains q1, q2, …, q23. b. Applying the dual effect, we form another 23 × 23 CTC(initial dual) matrix from CTC(initial). In this new matrix, its entries of [๐(i, j)], for i = 1, 2, … , 23 and j = 24, 25, … , 46, are the first 23 roots starting from 11 for “i”s and the 24th to 46th roots for “j”s, respectively. Namely, the left column contains r1, r2, …, r23, and the top row contains q24, q25, …, q46. c. Add r47 as the end of the 24th row and add q47 and q48 as the end of the 47th and the 48th columns. Namely, the left column contains r47 as the last index and the top row contains q47, q48 as the last two indices. d. Combine matrices developed by a, b, and c steps to complete CTC(basic). As shown in Figure 1, there are the CTC(initial), the CTC(initial-dual), the CTC(basic), the CTC(complementary), and CTC(1). CTC (initial) = {๐(i, j) | i = 1, 2, โฏ , 23; j = 1, 2, โฏ , 23}, CTC (initial-dual) = {๐(i, j) | i = 1, 2, โฏ , 23; j = 24, โฏ , 46}, CTC(basic) = {๐(i, j) | i = 1, 2, โฏ , 23, 47; j = 1, 2, โฏ , 48}, CTC(complementary) = {๐(i, j) | i = 24, 25, โฏ , 46, 48; j = 1, 2, โฏ , 48}, CTC(1) = {๐(i, j) | i, j = 1, 2, โฏ , 48}. 2. Example for Table S1(a) Given q1 = 11, compute ๐(1, 1), as follows: q ×q ๐ก(๐ข,๐ฃ) โต ๐(1, 1) = 1 + 1 210 q ๐ก(๐ข,๐ฃ) = 211, ∴ ๐(1, 1) = 1 + −r1 11×211−11 210 , and the unique q ๐ก(๐ข,๐ฃ) to let 11×q๐ก(๐ข,๐ฃ) −11 210 be an integer is = 1 + 11 = 12. Similarly, we find ๐(2, 1), ๐(3, 1) ๐(4, 1), ๐(5, 1), ๐(6, 1), โฏ , ๐(23, 1), ๐(47, 1) = (10, 6, 4, 11, 5, โฏ , 8, 1). 3. Elaboration of the Dual Effect of the CTC(basic) Electronic copy available at: https://ssrn.com/abstract=4742238 15 In the CTC(1), the dual effect made in Statement 2 can be further elaborated below between ๐(๐ข, ๐ฃ) and ๐(๐ข, ๐ก(๐ข, ๐ฃ)). (i) For ๐(๐ข, ๐ฃ) = ๐(๐ข, ๐ก(๐ข, ๐ฃ)), ๐ข๐ ๐ ≤ ๐ฃ, ๐ก(๐ข, ๐ฃ) ≤ ๐๐, ๐ญ๐ก๐๐ง ๐(๐ข, ๐๐ − ๐ฃ) = ๐(๐ข, ๐๐ − ๐ก(๐ข, ๐ฃ)) ๐ช๐๐−๐ฃ × ๐ช๐๐−๐ก(๐ข,๐ฃ) ๐ซ๐ ๐ซ๐ =๐+ ( + ๐(๐ข, ๐ฃ) − ๐) − . ๐ช๐ฃ × ๐ช๐ก(๐ข,๐ฃ) ๐๐๐ ๐๐๐ (ii) For ๐(๐ข, ๐ฃ) = ๐(๐ข, ๐ก(๐ข, ๐ฃ)), ๐๐จ๐ซ ๐ ≤ ๐ฃ ≤ ๐๐ ๐๐ฎ๐ญ ๐ง๐จ๐ญ ๐๐จ๐ซ ๐ก(๐ข, ๐ฃ), then ๐๐ ≤ ๐ก(๐ข, ๐ฃ) ≤ ๐ซ +๐๐๐[๐(๐ข,๐ฃ)−๐] ๐๐ ๐๐ง๐ ๐ช๐ก(๐ข,๐ฃ) = ๐ ๐ช๐ฃ . By utilizing Statement 2, we can generate a new matrix CTC(initial dual) in the Introduction from CTC(initial). Taking the first row, i.e. i = 1 and r1 = 11 as an example (Figure 2): (i) Since ๐(1, 2) = ๐(1, 3) = 2, we have ๐(1, 47 − 2) = ๐(1, 47 − 3) = 1 + q 45 × q 44 11 11 ( + ๐(1, 2) − 1) − = 182 q 2 × q 3 210 210 Similarly, since ๐(1, 5) = ๐(1, 8) = 5, we have q 42 × q 39 11 11 ๐(1, 47 − 5) = ๐(1, 47 − 8) = 1 + ( + ๐(1, 5) − 1) − = 155 q 5 × q 8 210 210 (ii) For i = 1 and j = 1, 6, 10, 12, โฏ , 19, 21, 22, we do not have 1 ≤ h(i, j) ≤ 23 to fit ๐(1, j) = ๐(1, h(i, j)), then we compute ๐(1, h(i, j)) for 24 ≤ h(i, j) ≤ 46. When j = 1, q1 = 11 and ๐(1, 1) = 12, since r1 +210(๐(1,1)−1) q1 = 11+210(12−1) 11 = 211 = q 48 , we have h(1, 1) = 48 and ๐(1, 48) = ๐(1, 1) = 12 When j = 6, q 6 = 29 and ๐(1, 6) = 16, r +210(๐(1,6)−1) since 1 q6 = 109 = q 25 , we have h(1, 6) =25 and ๐(1, 25) = ๐(1, 6) = 16. Similarly, we can predict ๐(1, 26) = ๐(1, 15) = 37, ๐(1, 28) = ๐(1, 12) = 33, ๐(1, 29) = ๐(1, 14) = 39, ๐(1, 30) = ๐(1, 10) = 29, โฏ , ๐(1, 41) = ๐(1, 22) = 88. 4. An Associated Example for Statement 3 Electronic copy available at: https://ssrn.com/abstract=4742238 16 Comparing the CTC(basic) with the CTC(complementary) and referring to Table S1(a) and Table S1(b), it is clear that: Given j = 1 and q1 = 11, we have ๐(1, 1) + ๐(46, 1) = 12 + 11 = 2q1 + 1. Given j = 2 and q2 = 13, we have ๐(2, 2) + ๐(45, 2) = 14 + 13 = 2q 2 + 1. Given j = 1 and q1 = 11, for i = 2, we have ๐(45, 1) + ๐(2, 1) = 2 + 10 = 12 = q1 + 1. Given j = 12 and q12 = 53, for i = 1, ๐(46, 12) + ๐(1, 12) = 21 + 33 = 54 = q12 + 1; for i = 23, ๐(24, 12) + ๐(23, 12) = 28 + 26 = 54 = q12 + 1; for i=47, ๐(47, 12) + ๐(48, 12) = 26 + 27 = 53 = q12. 5. Statement and proof of the intermedia effect in the CTC(1) Table S1(a) and Table S1(b) show another distribution of ๐(i, j) on the CTC, named the intermedia effect, which happens at some symmetrical columns, described below. The intermedia effect in the CTC(1) In CTC(1), there exists the intermedia effect between columns 23 and 24 such that, for ๐ข = ๐, ๐, … , ๐๐, (i) (ii) if ๐(๐ข, ๐๐) + ๐(๐ข, ๐๐) = ๐๐๐, ๐ญ๐ก๐๐ง ๐(๐ข, ๐๐) = ๐(๐๐ − ๐ข, ๐๐), if ๐(๐ข, ๐๐) + ๐(๐ข, ๐๐) = ๐๐๐, ๐ญ๐ก๐๐ง ๐(๐ข, ๐๐) + ๐ = ๐(๐๐ − ๐ข, ๐๐). For instances, in the case of green columns 23 and 24 of Figure 2, ๐(19, 23) = ๐(28, 24) = 6, โฏ , ๐(2, 23) + 2 = ๐(45, 24) = 77. Other column pairs also exhibit the intermedia effect. For example, in the case of columns 22 and 25, if ๐(i, 22) + ๐(i, 25) = 110, then ๐(i, 22) = ๐(47 − i, 25). Proof of the intermedia effect in the CTC For the case of columns 23 and 24, (i) From the mirror effect, we have ๐(i, 24) + ๐(47 − i, 24) = q 24 + 1 = 108. If ๐(i, 23) + ๐(i, 24) = 108, then ๐(i, 23) = 108 − ๐(i, 24) = ๐(47 − i, 24). (ii) If ๐(i, 23) + ๐(i, 24) = 106, then ๐(i, 23) + 2 = 108 − ๐(i, 24) = ๐(47 − i, 24). For the case of columns 22 and 25, from the mirror effect, we know ๐(i, 25) + ๐(47 − i, 25) = q 25 + 1 = 109 + 1 = 110 If ๐(i, 22) + ๐(i, 25) = 110, then ๐(i, 22) = ๐(47 − i, 25). 6. Numerical Illustrations for Statement 4 Take θ = 2 (i.e., cycle 2) as an example. CTC(2) consisting of CTC(basic) and CTC(complementary) is presented in Table S2(a) and Table S2(b), respectively. Table S2(a) is for ๐(i, j + 48) where 1 ≤ i ≤ 23. Table S2(b) is for ๐(i, j + 48), where 24 ≤ i ≤ 46 and 47, 48. Electronic copy available at: https://ssrn.com/abstract=4742238 17 Table S2(a) can be deducted from the CTC(basic) of Table S1(a) by the cyclic effect of Statement 4. For instance, ๐(i, 49) in Table S2(a), for 1 ≤ i ≤ 23, is computed as follows ๐(i, 49) = ๐(i, 1) + (2 − 1)q ๐ก(๐ข,๐ฃ) = ๐(i, 1) + r๐ + 210(๐(i, j) − 1) q1 Since (๐(1,1), ๐(2,1), ๐(3,1), ๐(4,1), โฏ , ๐(23,1)) = (12, 10, 6, 4, โฏ , 8), and q1 = 11, we have (๐(1,49), ๐(2,49), ๐(3,49), โฏ , ๐(23,49)) = (223, 183, 103, 63, โฏ, 151). Similarly, since (๐(1,2), ๐(2,2), ๐(3,2), ๐(4,2), โฏ , ๐(23,2)) = (2, 14, 12, 11, โฏ , 8), and q1 = 11, we then have (๐(1,50), ๐(2,50), ๐(3,50), ๐(4,50), โฏ , ๐(23,50)) = (19, 225, 191, 174, โฏ , 129). Table S2(b) is deduced from Table S2(a) by the mirror effect of Statement 3. For instance, for q 49 = 221, we have ๐(24, 49) = q 49 + 1 − ๐(23, 49) = 221 + 1 − 151 = 71, and ๐(46, 49) = 2q 49 + 1 − ๐(1, 49) = 442 + 1 −223 = 220. Therefore, using the CTC(1), we can generate the CTC(θ), for θ = 2, 3, … For instance, given r11 = 11, we have ๐(1,1) = l + 211, and ๐(1, 1) = 12. 11×q๐ก(๐ข,๐ฃ) −11 210 . Therefore, q ๐ก(๐ข,๐ฃ) = For θ = 2, we have ๐(1, 49) = 12 + (2 − 1) × 211 = 223. For θ = 3, we have ๐(1, 97) = 12 + (3 − 1) × 211 = 434. For θ = 4, we have ๐(1, 145) = 12 + (4 − 1) × 211 = 645. Similarly, ๐(1, 2) = 2. Therefore, we find ๐(1, 49) = 2 + (2 − 1) × 211 = 213 and ๐(1, 98) = 2 + (3 − 1) × 211 = 424. 7. Proofs of the Statements Proof of Statement 1 For each i, j = 1, 2, โฏ 48, notice that in CTC(1), ๐(j, j) = 1 + q j × q ๐ก(๐ฃ,๐ฃ) − rj 210 =1+ q j (211 − 1) 210 = 1 + q j , as rj = q j and we choose q ๐ก(๐ฃ,๐ฃ) = q 48 = 211. For a given j, ri = q jฬ × q ๐ก(๐ข,๐ฃ) − (๐(i, j) − 1) × 210, for some h(i, j). This means that a Electronic copy available at: https://ssrn.com/abstract=4742238 18 fixed ri is the remainder of q jฬ × q h(i,j) subtracting the maximum multiple of 210 smaller than q j × q ๐ก(๐ข,๐ฃ) . Since q h(j,j) = q 48 = 211 is the largest q ๐ก(๐ข,๐ฃ) hence ๐(j, j) ≥ ๐(i, j). Proof of Statement 2 By definition, ๐(i, j) = 1 + qj qh(i,j) − ri 210 q q − ri , ๐(i, h(i, j)) = 1 + ๐ก(๐ข,๐ฃ)210๐ฃ . q q๐ก(๐ข,๐ฃ) − ri Therefore, we have ๐(i, j) = ๐(i, h(i, j)) = 1 + j× 210 Proof of Statement 3 (i): Notice that r47−i + ri = 210, ∀ i = 1, โฏ , 23, q q๐ก(๐ข,๐ฃ) −r47−i ๐(47 − i, j) = 1 + j ๐(i, j) = 1 + 210 qj q๐ก′(๐ข,๐ฃ) −ri 210 for some q ๐ก′(๐ข,๐ฃ) and for some q ๐ก′(๐ข,๐ฃ) . Hence, q j (q ๐ก(๐ข,๐ฃ) + q ๐ก′(๐ข,๐ฃ) ) − 210 210 q j (q ๐ก(๐ข,๐ฃ) + q ๐ก′(๐ข,๐ฃ) ) =1+ 210 ๐(47 − i, j) + ๐(i, j) Since =2+ qj (q๐ก(๐ข,๐ฃ) +q๐ก′(๐ข,๐ฃ) ) 210 must be an integer, and q j has no factors of 2, 3, 5, 7, q ๐ก′(๐ข,๐ฃ) + q ๐ก(๐ข,๐ฃ) must be a multiple of 210. This happens only when q ๐ก(๐ข,๐ฃ) + q ๐ก′(๐ข,๐ฃ) = { 210, if h(i, j) + h′ (i, j) = 47 420, if h(i, j) or h′ (i, j) = 48 Therefore, ๐(47 − i, j) + ๐(i, j) = { (ii): q j + 1, if i ≠ j 2q j + 1, if i = j Notice that r47 + r48 = 209 + 211 = 420. Hence q q๐ก(๐ข,๐ฃ) −r47 j ๐(47, j) + ๐(48, j) = (1 + 210 = q q๐ก′(๐ข,๐ฃ) −r48 ) + (1 + j 210 ) qj (q๐ก(๐ข,๐ฃ) +q๐ก′(๐ข,๐ฃ) ) 210 q j , if h(i, j) + h′ (i, j) = 47 ={ 2q j , if h(i, j) or h′(i, j) = 48 The 2nd case happens only when j = 47, 48. Electronic copy available at: https://ssrn.com/abstract=4742238 19 Proof of Statement 4 ๐(i, j) + (θ − 1)q ๐ก(๐ข,๐ฃ) = 1 + q j q ๐ก(๐ข,๐ฃ) − ri + 210(θ − 1)q ๐ก(๐ข,๐ฃ) 210 [q j + 210(θ − 1)]q ๐ก(๐ข,๐ฃ) − ri 210 q j+48(θ−1) q ๐ก(๐ข,๐ฃ) − ri =1+ 210 =1+ = ๐(i, j + 48(θ − 1)) 8. Justification for Establishing ๐๐ (๐) and Table S3 Starting with θ = 1, compute q j , q h(1,j) and ๐(1, j) using Expression (1) q j × q ๐ก(๐ข,๐ฃ) − 11 ๐(1, j) = 1 + , for j = 1, 2, โฏ , 48. 210 Then by the cyclic effect of Statement 4, ๐(1, j + 48(θ − 1))s, for 2 ≤ θ ≤ 19, are computed based on ๐(1, j): ๐(1, j + 48) = ๐(1, j) + ๐๐ก(๐,๐ฃ) ๐(1, j + 96) = ๐(1, j) + 2๐๐ก(๐,๐ฃ) โฎ ๐(1, j + 864) = ๐(1, j) + 18๐๐ก(๐,๐ฃ) We then list ๐(i, j + 48(θ − 1)) for θ = 1, 2, 3, … , 19 in Table S3. The bottom row of Table 3S shows that โ1,1 = 2, โ1,2 = 15, โ1,3 = 28, โฏ , โ1,19 = 210. Notably, min{๐(i, j + 48(θ − 1))} < min{๐(i, j + 48θ)} for all θ = 1, 2, 3, … , 19, and โi,1 < โi,2 < j j โi,3 < โฏ , for all i. For instance, given b = 2,311 = 211 + 210 × 10, compute L2,311 (1) by applying Expression (2). Referring to Expression (2), Lb (i) is a set consisting of ๐(i, j) ๏ฃ โi,θ+1. Since โ๐,θ ≥ 2311 ⌈ 211 ⌉ = 11, choose θ = 2 to satisfy โi,θ = min{๐(1, j + 48)} = 15. We then obtain j L2311 (1) = {2, 5, 7, 9, 12, 15}. According to the Formula of Primes, in order for α = 11 + 210k ≤ 2311 to be primes, it requires k + 1 ∉ L2311 (1) for k ∈ ๏ป1, 2, … , 10๏ฝ. Therefore, we let k ∈ {0, 2, 3, 5, 7, 9, 10} and get the following primes: 11, 11 + 210 × 2 = 431, 11 + 210 × 3 = 641, 11 + 210 × 5 = 1061, 11 + 210 × 7 = 1481, 11 + 210 × 9 = 1901 and 11 + 210 × 10 = 2111. They are all shown in the first row of Table 1. Given b = 211 + 210 × 28 = 6091, we choose θ = 3 to satisfy โi,θ = min{๐(1, j + 96)} = 28, then L6091 (1) = j {2, 5, 7, 9, 12, 15, 16, 17, 19, 23, 24, 28}. Electronic copy available at: https://ssrn.com/abstract=4742238 20 In order for α = 11 + 210k ≤ 6091 to be primes, we let k ∈ {0, 2, 3, 5, 7, 9, 10, 12, 13, 17, 19, 20, 21, 24, 25, 26} such that k+1∈ L6091 (1) to obtain the following primes: 11, 431, 641, 1,061, 1,481, 1,901, 2,111 2,531, 2,741, 3,581, 4,001, 4,211, 4,421, 5,051, 5,261, 5,471. Similarly, we can list all Lb (i) and all primes α = 11 + 210k ≤ b for a given b. 9. Proofs of Expression (3) and Expression (4) Proof of Expression (3) The total number of elements in PTP + (0, k) is 48 × (k + 1), for k ∈ N0 . Within the interval [11, 2112 θ], the number of different ๐(i, j) is |Lb (i)| for b = 211 + 210k. Hence, the total number of composites without 2, 3, 5, and 7 within the above interval is |Lb (i)|. Note that 4 primes are less than 10. Therefore, the number of primes within [2, b] is π(b) = 4 + 48 × (k + 1) − ∑48 ๐=1|Lb (i)|. Proof of Expression (4) For an integer i ∈ W = {1, 3, 6, 9, 13, โฏ , 45, 47}, (r๐ , r๐+1 ) are twin primes for r๐ ≥ 11. If there exists a k, k ∈ ๐0, such that k + 1 ∉ Lb (i) and k + 1 ∉ Lb (i + 1), then r๐ + 210k and r๐+1 + 210k are twin primes. Since (2, 3), (3, 5), (5, 7) are the only three twin-prime pairs less than 11. Therefore, the number of twin-prime pair in interval [2, b] is π∗ (b) = 3 + ∑๐∈W|T(i, i + 1)|, where T(i, i + 1) is a set of k such that k + 1 ∉ Lb (i) and k + 1 ∉ Lb (i + 1). 10. Numerical Illustrations for The Periodic Table of Primes Let λ(i, k) be the integer value of the entry (i, k) in PTP+ (0, 210) if Figure 3, which is computed based on CTC (1) and the Formula of Primes. For instance, λ(i, k) on the diagonal dotted points (1, 11), (2, 13), (3,17), (4, 19) and (5, 23) are computed below. ๐(1, 1) = 12, ๐(2, 2) = 14, โฏ, ๐(23, 23) = 104, โฏ, ๐(46, 46) = 200 are converted to λ(1, 11), λ(2, 13), … λ(23, 17), โฏ, λ(46, 103) , respectively, shown by red numbers as λ(1, 11) = r1 + 210(๐(1, 1) − 1) = 11 + 210(12 − 1) = 2321 = 11 × 211, λ(2, 13) = r2 + 210(๐(2, 2) − 1) = 13 + 210(14 − 1) = 2743 = 13 × 211, โฎ Similarly, λ(23, 103) = r23 + 210(๐(23, 23) − 1) = 103 + 210(104 − 1) = 21733 = 103 × 211. λ(46, 199) = r46 + 210(๐(46, 46) − 1) = 199 + 210(200 − 1) = 41989 = 199 × 211. Also, ๐(46, 1) = 11, ๐(45, 2) = 13, โฏ, ๐(1, 46) = 199 are converted to λ(i, k), respectively, as λ(46, 10) = r1 + 210(๐(46, 1) − 1) = 199 + 210(11 − 1) = 2299 = 11 × 209 λ(45, 12) = r2 + 210(๐(45, 2) − 1) = 197 + 210(13 − 1) = 2717 = 13 × 209 Electronic copy available at: https://ssrn.com/abstract=4742238 21 โฎ λ(1, 210) = r46 + 210(๐(1, 46) − 1) = 11 + 210(199 − 1) = 41591 = 199 × 209. Now return to columns 22 and 25 in Figure 2. We note that (i, 23) and (i,24), for i = 1, 2, …, 48 being the intermediate dotted points. There are two other green lines 12–10–9–7–6 for column 22, and 6–7–9–10–12 for column 25. Both are caused by the intermedia effect, with respect to 47. These two lines are also converted to Table 1, shown as 11×101–13×101–17×101–19×101–23×101, and 11×109– 13×109–17×109–19×109–23×109, respectively. The above conversions between ๐(i, j) in the CTC and λ(i, k) in the PTP+ demonstrate that (i) many composties such as the digonal and intermediate dotted points which allocate cyclically at the CTC are also allocatted periodically at the PTP+. (ii) the PTP is obtained from removing all composites periodically allocated at PTP+. Therefore, we claim that prime numbers are also generated periodically since it comes from removing all periodically distributed composites. Table 1 contains the first 11 columns for the PTP+(0, 210), which is composed of 48 × 11 = 528 elements, represented by λ(i, k) and elaborated below. (i) Each entry λ(i, k) which is either a prime or a composite. (ii) If λ(i, k) is a prime, then λ(i, k) = r๐ + 210k , and vice versa. If λ(i, k) is a composite, then there are q j and q (i,j) such that λ(i, k) = q j × q ๐ก(๐ข,๐ค) = ri + 210 × (๐(i, k + 1) − 1). Take i = 1, r๐ = 11 for instance, λ(1, 1) = 11 + 210 × (๐(1, 2) − 1) = q 2 × q 3 = 13 × 17, λ(1, 4) = 11 + 210 × (๐(1, 5) − 1) = q 5 × q 8 = 23 × 37. Then, we have λ(1, 0) = 11, λ(1, 2) = 11 × 210(3 − 1) = 431, λ(1, 3) = 11 + 210 × (4 − 1) = 641, โฏ, λ(1, 9) = 11 + 210 × (10 − 1) = 1,901. (iii) Primes λ(i, k) and λ′(i, k) = λ(i, k) + 1 are twin-prime pairs, if and only if r๐′ = 2 + r๐ , where λ(i, k) = r๐ + 210 × k, λ′ (i, k) = r๐′ + 210 × k. For instance, (431, 433), (641, 643), (1,061, 1,063), (1,481, 1,483) are twinprime pairs. (iv) The numbers of primes in column k of the PTP+ (0, 210) are shown as (43, 35, 32, 31, 31, 28, 28, 30, 26, 27, โฏ ). The sum of primes within interval [11, 2,311] is 43 + 35 + 32 + โฏ + 18 = 329. The number of twin-prime pairs of a column k are listed as (12, 6, 4, 4, 4, 8, 6, 4, 4, 4, โฏ , 3). The maximum prime gap for the kth column can also be found from Table 1. For instance, the maximum prime gap for column 2 is 10, occurring at 409 and 419 of r46 and r47 , respectively. Electronic copy available at: https://ssrn.com/abstract=4742238 22 (v) The PTP(k0, k’) is useful in predicting primes. Examples are given by Nine Examples for the Predictions of Primes by the PTP. In summary, the Formula of Primes is checked below: From Table S1(a), we know that Lb (1) = {2, 5, 7, 9, 12, โฏ } Lb (2) = {7, 10, 12, 13, 14, โฏ } Lb (3) = {3, 6, 11, 12, 13, 14, โฏ } โฎ From Table 1, we know that α = r1 + 210k is prime for k = 0, 2, 3, 5, 7, 9, 10 α = r2 + 210k is prime for k = 0, 1, 2, 3, 4, 5, 7, 8, 10 α = r3 + 210k is prime for k = 0, 1, 3, 4, 6, 7, 8, 9 โฎ In general, α = r๐ + 210k is prime, if k + 1 ∉ Lb (i) 11. Nine Examples for the Predictions of Primes by the PTP Example 1: Prediction of the next prime after 1,951 Since 1951 = 61 + 210(10 − 1), from Table 1 there are four consecutive composites, i.e. 19 × 103, 37 × 53, 13 × 151, and 11 × 179, right after 1,951. We know the prime next to 1,951 is 1,973. Example 2: Prediction of the 300th prime From Expression (3), given b = 1,891, the number of primes within ๏2, 1,891๏ is counted as 48 π(1,891) = 4 + 48 × (8 + 1) − ∑|Lb (i)| = 4 + 432 − 152 = 284. ๐=1 The 284th prime is 1889, which is located at (i, k) = (47, 8) as shown in Table 1. Therefore, the 300th prime is counted to be 1,993, at the location of (i, k) = (23, 9). Example 3: Prediction of the next twin prime after the twin-prime pair (1277, 1279) The location of 1,277 is at (i, k) = (3, 6). The twin prime λ(i, k) are and λ′ = ๐(i, k) + 1 expressed as λ(i, k) = r๐ + 210(7 − 1). λ′ = r๐′ + 210(7 − 1) and r๐′ = r๐ + 2. Therefore, r๐ = 29, r๐′ = 31, and λ(i, k) = 1,289, λ′ = 1,291. Electronic copy available at: https://ssrn.com/abstract=4742238 23 Example 4: Predicting the longest same-difference primes We can predict the longest same-difference prime series in PTP+(0, 10) from the CTC(basic). Checking row 1 of Table S1(a), L2311 (1, j) = {๐(1, 1), ๐(1, 2), ๐(1, 3), โฏ , ๐(1, 10), ๐(1, 11)} = {2, 5, 7, 9, 12, 16, 24, 29} By the mirror effect made in Statement 2 and via L2311 (1, j), L2311 (46, j) = {11, 12, 14, 15, 16, 19, 21, 24, 25, 33, 35, 15, 24} implies that 1, 2, 3, 4, 5, 6, 7, 8, 9, 10 do not belong to L2,311 (46, j). As shown on row 46 of Table S1(b). Therefore, α is prime, for α = 199 + 210k, k = {0, 1, 2, 3, 4, โฏ , 9}. We then find the longest same-difference prime series in PTP+ [0, 10], or a 10-tuple primes series, to be (199, 409, 619, 829, 1,039, 1,249, 1,459, 1,669, 1,879, 2,089), as shown on row 46 of Table 1. Similarly, we can find the next longest same-difference prime series in PTP+ (0, 10). That happens at row 9, listed to be (881, 1,091, 1,301, 1,511, 1,721, 1,931, 2,141), as shown on row 9 of Table 1. Example 5: Visualization of the fundamental theorem of arithmetic How to factorize 46189 on CTC(2) and to visualize 46,189 on PTP+? Since 46,189=199+210 × 219, we find that the root of 46,189 is r46 = 199. Let 46,189 = q j × (q ๐ก(๐ข,๐ฃ) + 210(θ − 1)). We need to find ๐(46, j) = 1 + qj ×q๐ก(๐ข,๐ฃ) −199 210 . We find that a feasible solution to satisfy the above equation occurs when j = 32 and h(i, j) = 26. Namely, ๐(46, 32) = ๐(46, 26) = 77, with q 32 = 143 and q 26 = 113. Since 46,189 = q j × (q ๐ก(๐ข,๐ฃ) + 210(θ − 1)) with θ = 2, we have 46189 = 143 × (113 + 210) = 143 × 323 = 11 × 13 × 323. On CTC (2), we find q 32 and q (26+48) = q 74 to fit 46,189 = q 32 × q 74, as shown in Figure S1(a). On PTP+ (0, 219), we have (i, k) = (1, 74) and (i′ , k ′ ) = (1, 32), where λ(1, 74) = 323 and Electronic copy available at: https://ssrn.com/abstract=4742238 24 λ(1, 32) = 143 such that 46189 = λ(1, 74) × λ(1, 32). Visualization of CTC(2) and PTP+(0, 219) can be seen in Figure S1(b). Example 6: The maximum prime gap in the interval [10,000, 300,000] The prime gap in the interval of two primes ๏๏กi, ๏กj ๏ is ๏กj - ๏กi, where (i) (ii) (iii) ๏กi = ri + 210k is a prime rooted to ri ∈ S, and k+1 ∉ Lb (i), ๏กj = ri+n + 210k is a primed rooted to ri+n ∈ S for some n and k + 1 ∉ Lb (i + n), and ๏กi+m = ri+m + 210k is a composite for m = 1, 2, …, n-1 and k +1 ∈ Lb (i + m). In this way, checking L๐ (i) for b = 299,881 with b – 211 being a multiplier of 210, we find that i = 7 and k = 1,340 such that (i) (ii) (iii) ๏ก7 = r7 + 210 ×1,340 = 281,431 is a prime, where 1,340+1 = 1,341 ∉ Lb (7), ๏ก25 = r25 + 210 ×1,340 = 281,509 is a prime, where 1,340 +1 = 1,341∉ Lb (25), and ๏ก7+m = r7+m + 210×1,340 are composites for m = 1, 2, …, 17, where 1,340 +1 ∈ Lb (7 + m). Therefore, the prime gap within ๏281,431, 281,509๏ is 281,509 – 281,431 = 78, which is the maximum prime gap within ๏10,000, 300,000๏, where 281,431 is located at the entry (25, 1,340) at PTP+(47, 1,427) on Figure S2. Example 7: The largest twin-prime pair in the interval [10,000, 300,000] We can count the largest twin-prime gap using Expression (4). Checking PTP+ (47, 1,427), we find it to be (299,699, 299,701), where 299,699 = 29 + 210 × 1,427 and 299,699 = 31 + 210 × 1,427, which are located at (i, k) = (6, 1,427) and (7, 1,427), respectively. See Figure S2. Example 8: How many primes are within the interval [2, 2,311]? From the CTCs, we count to find ∑48 ๐=1|Lb (i)| = 188 for b =2,311. From Expression (3), π(2311) = 4 + 48(10 − 1) − 188 = 248. Electronic copy available at: https://ssrn.com/abstract=4742238 25 Example 9: How many twin-prime pairs are within the interval [11, 2311]? Referring to Expression (4), let b = 2311, from CTC, we compute T(i, i + 1), for i ∈ {1, 3, 6, 9, 13, 16, 22, 24, 30, 33, 37 40, 43, 45, 47} to reach T(1,2) = {k | k + 1 ∉ Lb (1), k + 1 ∉ Lb (2)} = {0, 2, 3, 5, 7,10} Similarly, we have T(3, 4) = {0, 1, 4, 6, 7, 8}, I(6, 7) = {0, 1, 3, 6}, and T(9, 10) = {0, 2, 4, 5, 6, 7, 8, 9}, and T(47, 48) = {0, 1, 4, 10}. Therefore, the number of twin-prime pairs is π∗ (2311) = 3 + ∑i∈w|T(i, i + 1)| = |T(1, 2)| + |T(3, 4)| + |T(6, 7)| + โฏ + |T(47, 48)| = 3 + 6 + 6 + โฏ + 4 = 69. 12. Justification of Claim 4 in Discussion Given a set of Lb (i, j + 48(θ − 1)), if there exist ๐(i, j∗ ) and ๐(i, j∗ + 1)with 11 ≤ j∗ ≤ 48θ such that ๐ (i, j∗ ), ๐ (i, j∗ + 1) ∈ Lb (i, j + 48(θ − 1)), for all 1 ≤ i ≤ 48, then the maximum prime gap within an interval [11, b] is 420. This implies that there exists no prime between α = r1 + 210 (๐(1, j∗ ) − 1) and α = r48 + 210 (๐(48, j∗ + 1) − 1). If there exist ๐(i, j∗ ), ๐(i, j∗ + 1), ๐(i, j∗ + 2) with 11 ≤ j∗ ≤ 48θ such that ๐(i, j∗ ), ๐(i, j∗ + 1), ๐(i, j∗ + 2) ∈ Lb (i, j + 48(θ − 1)) for all 1 ≤ i ≤ 48, then the maximum prime gap within an interval [11, b] is 630. This implies that there exists no prime between α = r1 + 210(๐(1, j∗ ) − 1) and α′ = r48 + 210 (๐(48, j∗ + 2) − 1). Electronic copy available at: https://ssrn.com/abstract=4742238 26 q๐ r๐ ๐(๐, ๐) r46 q26 q32 q74 =113 =143 =323 ๐(46, 26) ๐(46, 32) per r๐ k ๐(๐,k) r1 r2 r3 r4 0 26 32 143 74 r46 46189 (b) Figure S1: Visualization of 46189 and its factors root i r๐ period k 1 2 11 13 โฎ โฎ 6 7 29 31 โฎ 109 25 โฎ ๐(๐,k) max prime gap 47 โฏ 1340 โฏ 1427 299699 281431 219 323 โฎ ๐(46, 74) (a) iod 299701 281509 78 Figure S2: Maximum prime gap and the largest twin-prime pair in PTP[10000, 300000] Electronic copy available at: https://ssrn.com/abstract=4742238 27 Table S1(a): A basic composite table CTC(basic) j ent q segm root i 1 r๐ ๐(๐, ๐) 11 ๐ 1 2 3 4 5 6 7 8 9 10 11 12 11 13 17 19 23 29 31 37 41 43 47 53 โฏ 23 24 โฏ 46 47 48 โฏ 103 107 โฏ 199 209 211 12 2 2 9 5 16 7 5 7 29 24 33 2 13 10 14 13 7 12 24 25 36 23 38 41 34 24 84 199 12 75 31 197 14 3 17 6 12 18 3 3 11 30 24 14 13 28 4 19 4 11 12 20 10 19 17 18 30 22 45 36 74 32 193 18 37 22 86 191 20 5 23 11 9 17 16 24 6 22 6 21 40 32 39 21 87 187 24 6 29 5 6 16 10 22 30 14 25 28 24 36 42 71 35 181 30 7 31 3 5 10 8 6 9 32 19 3 33 6 43 19 89 179 32 8 37 8 2 9 2 4 4 24 38 10 17 10 46 69 37 173 38 9 41 4 13 14 17 18 20 29 26 42 35 44 48 68 38 169 42 10 43 2 12 8 15 2 28 16 20 17 44 14 49 16 92 167 44 11 47 9 10 13 11 16 15 21 8 8 19 48 52 15 93 163 48 12 53 3 7 12 5 14 10 13 27 15 3 5 54 65 41 157 54 13 59 8 4 11 18 12 5 5 9 22 30 9 4 12 96 151 60 14 61 6 3 10 16 19 13 23 3 38 39 26 5 63 43 149 62 15 67 11 13 4 10 17 8 15 22 4 23 30 8 10 98 143 68 16 71 7 11 9 6 8 24 20 10 36 41 17 10 9 99 139 72 17 73 5 10 3 4 15 3 7 4 11 7 34 11 60 46 137 74 18 79 10 7 2 17 13 27 30 23 18 34 38 14 7 101 131 80 19 83 6 5 7 13 4 14 4 11 9 9 25 16 6 102 127 84 20 89 11 2 6 7 2 9 27 30 16 36 29 19 56 50 121 90 21 97 3 11 16 18 7 12 6 6 39 29 3 23 54 52 113 98 22 101 10 9 4 14 21 28 11 31 30 4 37 25 53 53 109 102 23 103 8 8 15 12 5 7 29 25 5 13 7 26 104 107 107 104 47 209 1 7 3 1 8 25 22 3 33 17 15 26 26 210 210 Note: Numbers in red — for diagonal effect. Numbers in green — for intermediate effect. Electronic copy available at: https://ssrn.com/abstract=4742238 80 28 Table S1(b): A complementary composite table CTC (complementary) j ent q segm root i 24 r๐ ๐(๐, ๐) 107 ๐ 1 2 3 4 5 6 7 8 9 10 11 12 11 13 17 19 23 29 31 37 41 43 47 53 โฏ 23 24 โฏ 46 47 48 โฏ 103 107 โฏ 199 209 211 4 6 3 8 19 23 3 13 37 31 41 28 25 109 2 5 14 6 3 2 21 7 12 40 11 29 103 108 103 108 26 113 9 3 2 2 17 18 26 32 3 15 45 27 121 1 12 12 13 22 21 5 8 26 8 19 28 127 6 9 11 7 20 16 28 27 33 35 23 38 98 6 83 128 29 131 2 7 16 3 11 3 2 15 24 10 10 40 97 7 79 132 30 137 7 4 15 16 9 27 25 34 31 37 14 43 44 62 73 138 31 139 5 3 9 14 16 6 12 28 6 3 31 44 95 9 71 140 32 143 1 1 14 10 7 22 17 16 38 21 18 46 94 10 67 144 33 149 6 11 13 4 5 17 9 35 4 5 22 49 41 65 61 150 34 151 4 10 7 2 12 25 27 29 20 14 39 50 92 12 59 152 35 157 9 7 6 15 10 20 19 11 27 41 43 53 39 67 53 158 36 163 3 4 5 9 8 15 11 30 34 25 47 3 89 15 47 164 37 167 10 2 10 5 22 2 16 18 25 43 34 5 88 16 43 168 38 169 8 1 4 3 6 10 3 12 41 9 4 6 36 70 41 170 39 173 4 12 9 18 20 26 8 37 32 27 38 8 35 71 37 174 40 179 9 9 8 12 18 21 31 19 39 11 42 11 85 19 31 180 41 181 7 8 2 10 2 29 18 13 14 20 12 12 33 73 29 182 42 187 1 5 1 4 23 24 10 32 21 4 16 15 83 21 23 188 43 191 8 3 6 19 14 11 15 20 12 22 3 17 82 22 19 192 44 193 6 2 17 17 21 11 2 14 28 31 20 18 30 76 17 194 45 197 2 13 10 13 12 6 7 2 19 6 7 20 29 77 13 198 46 199 11 12 16 11 19 14 25 33 35 15 24 21 80 24 11 200 48 211 10 6 14 18 15 4 9 34 8 26 32 27 77 27 208 212 51 55 101 110 31 50 56 97 114 35 48 58 89 122 Note: Numbers in red — for diagonal effect. Numbers in green — for intermediate effect. Electronic copy available at: https://ssrn.com/abstract=4742238 29 Table S2(a): Upper rows of CTC(2) j ent q segm root i 1 r๐ ๐(๐, ๐) 11 ๐ 49 50 221 223 โฏ Table S2(b): Lower rows of CTC(2) 96 j ent q segm root i 24 r๐ ๐(๐, ๐) 107 ๐ 49 50 221 223 71 95 78 223 19 225 25 109 31 2 13 183 3 17 103 191 26 113 172 44 4 19 63 174 27 121 12 199 5 23 204 140 28 127 113 48 6 29 84 89 29 131 33 114 7 31 44 72 30 137 134 63 8 37 145 21 31 139 94 46 9 41 65 210 32 143 14 12 149 115 184 10 43 25 193 33 11 47 166 159 34 151 75 167 157 176 116 12 53 46 108 35 13 59 147 57 36 163 56 65 14 61 107 40 37 167 197 31 15 67 208 212 38 169 157 14 16 71 128 178 39 173 77 203 17 73 88 161 40 179 178 152 18 79 189 110 41 181 138 135 19 83 109 76 42 187 18 84 20 89 210 25 43 191 159 50 193 119 33 21 97 50 180 44 22 101 191 146 45 197 39 222 129 46 199 220 205 47 209 20 120 48 211 202 104 23 103 151 Electronic copy available at: https://ssrn.com/abstract=4742238 โฏ 96 30 Table S3: List of L44521((1, j + 48(θ-1)), for θ = 1, 2, …, 19 θ=1 j qj qh(๐,๐) 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 11 13 17 19 23 29 31 37 41 43 47 53 59 61 67 71 73 79 83 89 97 101 103 107 109 113 121 127 131 137 139 143 149 151 157 163 167 169 173 179 181 187 191 193 197 199 211 17 13 89 37 109 41 23 31 137 103 127 139 131 113 151 167 149 157 19 143 181 47 163 29 67 191 53 61 43 59 37 79 71 83 107 73 179 187 169 101 173 121 197 193 209 47 48 209 211 199 11 โ๐,θ ๐(1,j) 12 2 2 9 5 16 7 5 7 29 24 33 40 39 37 52 59 57 63 9 67 88 24 84 16 37 111 33 39 29 40 67 57 52 63 84 59 145 155 145 88 155 111 182 182 199 θ=2 θ=3 θ = 19 ๐(1, j+48) = ๐(1, j)+qh(๐,๐) ๐(1, j+96) = ๐(1, j)+2qh(๐,๐) ๐(1, j+864) = ๐(1, j)+18qh(๐,๐) 223 19 15 98 42 125 48 28 38 166 127 160 179 170 150 203 226 206 220 28 210 269 71 247 45 104 302 86 100 72 99 104 136 123 146 191 132 324 342 314 189 328 232 379 375 408 434 36 28 187 79 234 89 51 69 303 230 287 318 301 263 354 393 355 377 47 353 450 118 410 74 171 493 139 161 115 158 141 215 194 229 298 205 503 529 483 290 501 353 576 568 617 โฏ 3810 308 236 1611 671 1978 745 419 565 2495 1878 2319 2542 2397 2071 2770 3065 2739 2889 351 2641 3346 870 3018 502 1243 3549 987 1137 803 1102 733 1479 1330 1557 2010 1373 3367 3521 3187 1906 3269 2289 3728 3656 3961 199 12 398 23 597 34 3781 210 2 15 28 210 โ๐,θ=min{๐(i, j+48(θ-1)}, i=1 j Electronic copy available at: https://ssrn.com/abstract=4742238 31 Table S4: The Periodic Table of Primes of various intervals 97 root i 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 r๐ period k 11 13 17 19 23 29 31 37 41 43 47 53 59 61 67 71 73 79 83 89 97 101 103 107 109 113 0 11 13 17 19 23 29 31 37 41 43 47 53 59 61 67 71 73 79 83 89 97 101 103 107 109 113 11x11 11x11 127 131 137 139 127 131 137 139 11x13 11x13 149 151 157 163 167 149 151 157 163 167 13x13 13x13 173 179 181 173 179 181 11x17 11x17 191 193 197 199 191 193 197 199 11x19 11x19 211 211 99 100 101 102 103 104 105 106 โฏ 21011 21221 21851 21013 21433 22063 49 51 53 54 55 56 57 58 โฏ 96 20807 21017 21227 21647 22067 20389 20599 20809 21019 21649 21859 10091 10301 11981 19961 20183 20393 21023 11351 21863 22073 10093 10303 10513 11353 19963 20399 10723 22079 2 3 4 5 20611 6 10937 7 8210319 10 โฏ 48 11777 11987 21661 21871 10099 11779 20407 10729 10939 11149 21247 10103431 10313 11783 21881 22091 19973 20201 641 20411 10733 1061 1481 1901 11159 11789 19979 21673 22093 223 433 64310529 853 10739 1063 10949 1483 169311369 11579 9883 10111 10321 10531 11161 227 647 857 206271277 1487 1697 1907 214679887 10957 11587 21683 12007 229 439 859 1069 1279 1489 1699 21893 10331 11171 11801 12011 22109 19991 233 443 653 863 206391283 1493 1913 20219 20849 21059 21269 10333 11173 19993 239 449 659 20431 10753 170911383 11593 206411289 1499 21061 21481 22111 10337 11177 11597 11807 19997 241 661 1291 20857 21067 21277 214879901 10133457 10343 10973 11393 214919907 11813 877 1087 1297 20231 20441 21701 21911 10139 10979 251 46120233 10559 881 1091 1301 1511 172111399 1931 21493 20443 21283 22123 10141463 673 883 10771 20011 1093 1303 1523 1723 1933 21499 11821 22129 10357 10567 10987 11197 11617 11827 12037 257 467 677 887 1097 206631307 20873 21503 21713 22133 10151 10781 11831 12041 20021 263 683 1103 20879 152321089 173311411 11621 9923 20249 21929 10993 11833 20023 20887 21727 12043 21937 22147 269 479 1109 1319 1949 215179929 10159 10369 10789 11839 12049 20029 20261 206811321 153121101 271 691 1741 1951 215219931 10163 11003 11213 11423 11633 21943 22153 277 487 907 1117 1327 1747 21313 21523 10169491 70110589 20477 281 911 10799 20897 21107 21317 9941 21737 22157 10177 20269 10597 20047 283 1543 175311437 20479 1123 20899 21319 21529 21739 22159 10181499 10391 12071 20051 70910601 919 1129 1549 1759 21323 9949 20483 20693 20903 11443 11863 12073 293 503 155321121 1973 21751 21961 22171 11447 50920287 71910607 929 20707 11027 1559 1979 11657 11867 21757 10399 11239 307 727 937 1567 1777 21341 1987 9967 20921 10193 11243 21347 21557 21767 21977 20063 311 52120297 10613 941 1151 20507 207171361 1571 10831 11251 20071 313 523 733 20509 1153 1783 1993 215599973 20719 20929 21139 22189 10627 11047 11257 317 947 10837 1367 178711467 1997 11677 21143 21563 11887 21773 12097 22193 10211 11261 12101 73910631 1579 178911471 1999 11681 20939 21149 21569 10427 11057 11897 12107 743 20521 953 10847 1163 1583 2003 207311373 21991 10429 11059 20089 20947 21157 21577 21787 12109 21997 331 541 75110639 1171 1381 1801 2011 11689 10223547 10433 11273 11903 12113 20323 20743 21163 11483 22003 337 757 20533 967 10853 1597 2017 11279 20327 20747 11069 21377 11699 21587 11909 12119 761 971 10859 1181 1601 181111489 10651 20101 20749 11071 20959 21379 21589 21799 347 557 977 10861 1187 160721169 11491 2027 11701 10007 10657 10867 11287 11497 20107 20333 20963 21383 21803 22013 349 769 20543 207531399 1609 2029 10009 10243563 10453 20113 20549 20759 11083 22229 353 77310663 983 1193 161321179 182311503 21599 11923 10247 10457 11087 20117 20341 10667 20551 21391 21601 11927 359 569 1409 1619 2039 11717 10459 11299 20347 991 1201 21187 571 1621 1831 21397 11719 21817 22027 10253 10463 20123 20771 11093 20981 21821 12143 22031 367 577 787 997 10883 162721191 21401 21611 11933 10259 20129 20353 20563 10889 207731423 20983 21193 11519 373 1213 2053 21613 11939 12149 10891 11311 11941 22037 22247 21617 58720357 797 1217 1427 1637 1847 21407 11731 10037 10267 10477 11317 11527 12157 20359 10687 22039 379 1009 1429 10039 10271 10691 11321 207891433 21419 21839 12161 22259 383 59320369 1013 1223 2063 10273599 809 1019 10903 11113 20143 21001 21211 2069 11743 11953 21841 12163 22051 389 1229 1439 10487 20147 601 811 1021 1231 11117 1861 10909 11119 11329 11959 20149 397 607 1237 1447 1657 1867 10289 10499 10709 11549 11969 401 821 1031 1451 1871 2081 10061 10501 11131 11971 20161 613 82310711 1033 1453 1663 187311551 2083 617 827 1667 1877 2087 10067 409 619 829 1039 1249 1459 1669 1879 2089 10069 419 839 1049 1259 1889 2099 10079 PTP(0, 48) 421 631 1051 1471 20173 50 20177 ๐(๐,k) 98 20593 52 Electronic copy available at: https://ssrn.com/abstract=4742238 144 30047 30059 30071 30089 30091 30097 30103 30109 30113 30119 30133 30137 30139 โฏ 30161 30169 30181 30187 30197 30203 30211 30223 PTP(97, 144) 30241 PTP(49, 96)