Abstract:Let $b \ge 2$ be an integer. Not much is known on the representation in base $b$ of prime numbers or of numbers whose prime factors belong to a given, finite set. Among other results, we establish that any sufficiently large integer which is not a multiple of $b$ and has only small (in a suitable sense) prime factors has at least four nonzero digits in its representation in base $b$.
What problem does this paper attempt to address?
The problem that this paper attempts to solve is about the properties of the number of non - zero digits in the representation of integers with a finite number of prime factors in a specific number system. Specifically, the author studies whether, when all prime factors of an integer belong to a given finite set, these integers can have only a small number of non - zero digits in the representation of an integer base \(b\).
### Specific problem description
1. **Background and motivation**:
- For some prime numbers of specific forms (such as those in the form of \(2^n + 1\) or decimal numbers with all digits being 1), we don't know whether there are infinitely many such prime numbers.
- In recent years, some important progress has been made in the digital representation of prime numbers. For example, Mauduit and Rivat [16] proved that the sum of digits in the binary representation of prime numbers is uniformly distributed.
- Maynard [17] proved that for any digit \(d\), there are infinitely many prime numbers that do not contain this digit.
2. **Core problems of the research**:
- Research on the digital representation of integers with a finite number of prime factors in a specific number system \(b\).
- Of particular concern is: Are there large enough integers whose prime factors are very small (i.e., less than a certain bound) and have only a few non - zero digits in the representation of base \(b\)?
3. **Main results**:
- Theorem 1.1 shows that for any base \(b\), there are only a finite number of integers that are not divisible by \(b\) and have a given number of non - zero \(b\)-ary digits, and whose prime factors belong to a given finite set.
- Theorem 1.2 provides a more specific estimate, especially for the case of \(k = 3\), giving the growth rate of the largest prime factor with \(j\).
4. **Conclusion**:
- The author proves these results by applying classical tools in Diophantine approximation (such as Schmidt's subspace theorem and the theory of linear forms in logarithms).
- The results show that large enough integers cannot simultaneously have very small prime factors and a \(b\)-ary representation with a small number of non - zero digits.
### Summary of mathematical formulas
- Let \(n\) be a positive integer, and \(P[n]\) represents the largest prime factor of \(n\).
- Let \(S=\{q_1,q_2,\ldots,q_s\}\) be a finite set of prime numbers.
- \([n]_S := q_1^{r_1}q_2^{r_2}\cdots q_s^{r_s}\), where \(r_1,r_2,\ldots,r_s\) are non - negative integers, and \(M\) is an integer coprime to \(q_1,q_2,\ldots,q_s\).
- The sequence \((u_j^{(k)})_{j\geq1}\) is the sequence of all positive integers that are not divisible by \(b\) and have at most \(k\) non - zero \(b\)-ary digits, arranged in ascending order.
### Main theorems
#### Theorem 1.1
For integers \(b\geq2\), \(k\geq2\) and positive real number \(\varepsilon\), let \(S\) be a finite set of prime numbers. Then for sufficiently large \(j\), we have:
\[ [u_j^{(k)}]_S < (u_j^{(k)})^\varepsilon \]
In particular, the largest prime factor of \(u_j^{(k)}\) tends to infinity as \(j\) increases.
#### Theorem 1.2
For integers \(b\geq2\), let \(S\) be a finite set of prime numbers. Then there exist effectively computable positive numbers \(c_4\) and \(c_5\) such that for all \(j\geq c_5\), we have:
\[ [u_j^{(3)}]_S \leq (u_j^{(3)})^{1 - c_4} \]