Research notes on every problem, and a library of the papers behind them. Built from the open erdos repository.
Updated
Source. Robert I. Saye, On two conjectures concerning the ternary digits of powers of two, J. Integer Seq. 25 (2022), Article 22.3.4, 9 pp., as identified on the source card: Lemma 1, stated on p. 3, proved in Section 5, pp. 7--8.
Read depth. Claims checked: the statement and the notation it uses (Section 2, p. 2) were read clause by clause on the print. The proof was read but not checked step by step; nothing here is independently reviewed.
Statement
Notation (p. 2). For integers and a positive integer , means . For with ternary expansion , the -th ternary digit is , so is the least significant digit.
Lemma 1 (p. 3). Let be a positive integer and put . Then:
(i) is the least positive integer with ;
(ii) for , if then and differ by a multiple of ;
(iii) for ,
The paper notes (p. 3, footnote 1) that , so part (i) says that has the full order modulo . In part (iii) is or , so as runs over the st digit takes all three values while, by part (i), the last digits stay fixed; this is the step the paper's search uses (p. 3).
Proof pointer
Section 5 (pp. 7--8). The key fact, proved by induction on by cubing, is that . Part (i) follows by induction on , ruling out and as the order modulo ; part (ii) reduces to part (i) by cancelling a power of two, a unit modulo ; part (iii) follows by expanding modulo and multiplying by .
Dependencies
None outside the paper.
Bears on
- Problem 406: only as the tool behind the computer search on the main result's page. The lemma itself says nothing about which powers of two avoid the digit .