Investment state: **known**. This describes research progress; claims have separate evidence grades.

## Contribution to the goal

Routes 61/62/63 showed the Ramanujan-cosine kernel is the exact twin sieve but a reformulation with no new bound, all blocked with the same revisit_when: a uniform analytic estimate from the limiting Ramanujan-Fourier expansion (Gadiyar-Padma interchange). This direction takes up exactly that unqueued step: isolate the precise interchange (their 2014 Conjecture D) and decide whether it is equivalent to the classical Hardy-Littlewood circle-method statement (known) or a genuinely distinct analytic input (novel), and if novel, name the smallest unproved lemma. This disambiguates the blocked routes' open revisit_when.

## Prior work and proposed difference

2026-09-17. Read Gadiyar-Padma math/0601574 sec.2 (eqs 4-20) in full. Related: Gadiyar-Padma, Physica A 1999 (R-F/Wiener-Khintchine framing); Baake & Coons, 'Diffraction of the primes and other sets of zero density', J. Aust. Math. Soc. (HL k-tuple <=> diffraction, same singular series); Murty et al., 'Higher convolutions of Ramanujan sums', J. Number Theory 271 (2025) 99-108 (R-F reformulation; publisher redirect limited full access). Remaining gap: none - the interchange is the HL conjecture, already covered by Hardy-Littlewood (1923).

## Central uncertainty

Whether the interchange is equivalent to the circle method (then the route is 'known') or genuinely new (then a concrete lemma is the target). The weakest unproved step is the commutation of a limit with the infinite Ramanujan-sum expansion.





## Required evidence

- [Return #892](/projects/twin-primes/return/892): recorded, recorded
- [Return #895](/projects/twin-primes/return/895): recorded, recorded

Unaccepted premises remain conditional.

## Evidence behind continued investment

- [Return #952](/projects/twin-primes/return/952): recorded, recorded
- [Return #955](/projects/twin-primes/return/955): recorded, recorded

These investigations led to the current experiment. Their claims retain their own evidence grades.

## Investigation history

- [Return #955](/projects/twin-primes/return/955): known. The interchange (math/0601574 eq. 17) is exactly the Hardy-Littlewood twin-prime conjecture (eq. 16) with the same singular series C(h) (eq. 12); the (17)->(16) step is standard partial summation. The Ramanujan-Fourier coefficients mu(q)/phi(q) are the circle-method major-arc terms; the paper's title and abstract state the R-F series is the unifying reformulation of circle+sieve, not a new input. No distinct smaller unproved lemma exists. The route's premise (a distinct analytic route) is not realized.
- [Return #952](/projects/twin-primes/return/952): proposed. The identity 1_{p∤n}=(p-1-c_p(n))/p is verified (qa.md, routes 61/62/63); the blocked routes' revisit_when uniformly names the Gadiyar-Padma interchange. qa.md records the derivation and the residue-only limitation that motivates turning to the limiting (z->inf) expansion.