Even integers as sums of two sums of two squares with few prime factors

Let N ≡ 2 mod 4 be a sufficiently large integer. We show that the number of n ≤ N such that n and N − n are both sums of two squares and Ω(n(N − n)) ≤ 11 is at least (3/2) S(N) N / (log N)^2, where S(N) is the binary Goldbach singular series. With other parameters the method shows that at least (3/5) S(N) N / (log N)^2 such n satisfy max(Ω(n), Ω(N − n)) ≤ 8. The result extends to the principal form of each of the nine imaginary quadratic fields of class number one (discriminants −3, −4, −7, −8, −11, −19, −43, −67, −163). For even N, the bound is 11 when N is not divisible by the ramified prime (for −4: N ≡ 2 mod 4; for −8: v_2(N) ≤ 2; for −3: N ≡ 2 mod 3). In the remaining low-order cases the bound is 12 or 13, because there every representation must contain extra factors of the ramified prime. The sequence n(N − n) has level of distribution 1, but this level is shared by all sieve components. Sifting the inert primes up to N^(1/2) at level N lands exactly on the sieving limit s = 2 of the linear sieve, so they are sifted only up to N^(1/2 − δ). A survivor that is not a pair of represented integers has two large inert prime factors on the same side. These survivors are removed by a switching upper bound at the Bombieri–Vinogradov level. The main term and the switching term have the same local factors for every modulus, so the singular series cancels and the same certified numerical inequality applies to all nine discriminants. Richert's logarithmic weights control the number of prime factors jointly and on each side. The final numerical inequalities are verified with outward-rounded interval arithmetic. This record contains the paper (PDF and AMS-LaTeX source, 8 pages, 2 figures, 2 tables), the interval-arithmetic verification script certify.py, the local-condition analysis for the nine discriminants, the floating-point parameter scans, the figure script, and their outputs. This is a preprint. The analytic sieve lemmas have not yet been independently refereed. A web literature search found no overlapping result; a database search has not been done. Companion paper: R. Chen, On primes p for which N − p is an almost prime sum of two squares, and other class number one forms, preprint (2026), doi:10.5281/zenodo.23136112 Source code: https://github.com/Ruqing1963/bilateral-two-squares-sieve

Authors

Publication Details

Journal
Zenodo (CERN European Organization for Nuclear Research)
Published
2026-10-04
DOI
https://doi.org/10.5281/zenodo.23137626
Primary Topic
Analytic Number Theory Research
Type
preprint
Controls
|||
ALL TIME
JAN
FEB
MAR
APR
MAY
JUN
JUL
AUG
SEP
OCT
preprint

Even integers as sums of two sums of two squares with few prime factors

Ruqing Chen
Zenodo (CERN European Organization for Nuclear Research)
Analytic Number Theory Research
preprint

Even integers as sums of two sums of two squares with few prime factors

Ruqing Chen
preprint en

Abstract

Let N ≡ 2 mod 4 be a sufficiently large integer. We show that the number of n ≤ N such that n and N − n are both sums of two squares and Ω(n(N − n)) ≤ 11 is at least (3/2) S(N) N / (log N)^2, where S(N) is the binary Goldbach singular series. With other parameters the method shows that at least (3/5) S(N) N / (log N)^2 such n satisfy max(Ω(n), Ω(N − n)) ≤ 8. The result extends to the principal form of each of the nine imaginary quadratic fields of class number one (discriminants −3, −4, −7, −8, −11, −19, −43, −67, −163). For even N, the bound is 11 when N is not divisible by the ramified prime (for −4: N ≡ 2 mod 4; for −8: v_2(N) ≤ 2; for −3: N ≡ 2 mod 3). In the remaining low-order cases the bound is 12 or 13, because there every representation must contain extra factors of the ramified prime. The sequence n(N − n) has level of distribution 1, but this level is shared by all sieve components. Sifting the inert primes up to N^(1/2) at level N lands exactly on the sieving limit s = 2 of the linear sieve, so they are sifted only up to N^(1/2 − δ). A survivor that is not a pair of represented integers has two large inert prime factors on the same side. These survivors are removed by a switching upper bound at the Bombieri–Vinogradov level. The main term and the switching term have the same local factors for every modulus, so the singular series cancels and the same certified numerical inequality applies to all nine discriminants. Richert's logarithmic weights control the number of prime factors jointly and on each side. The final numerical inequalities are verified with outward-rounded interval arithmetic. This record contains the paper (PDF and AMS-LaTeX source, 8 pages, 2 figures, 2 tables), the interval-arithmetic verification script certify.py, the local-condition analysis for the nine discriminants, the floating-point parameter scans, the figure script, and their outputs. This is a preprint. The analytic sieve lemmas have not yet been independently refereed. A web literature search found no overlapping result; a database search has not been done. Companion paper: R. Chen, On primes p for which N − p is an almost prime sum of two squares, and other class number one forms, preprint (2026), doi:10.5281/zenodo.23136112 Source code: https://github.com/Ruqing1963/bilateral-two-squares-sieve

Zenodo (CERN European Organization for Nuclear Research)
Analytic Number Theory Research
AI Navigator

Ask Laika to Summarize, Analyze, and Connect papers live on the map.

Summarize Papers & Methodologies

Extract key findings, datasets, and comparative methods across publications.

Benchmark Rankings & Visual Analytics

Rank top research institutions, authors, funders, topics, and journals by Field-Weighted Citation Impact (FWCI) and paper volume with instant charts.

Connect Distant Disciplines

Bridge topological clusters on the map to find hidden collaborative intersections.

Even integers as sums of two sums of two squares with few prime factors — Ruqing Chen · Zenodo (CERN European Organization for Nuclear Research) (2026) | TGRS Research Map | TGRS