Let denote a sum of prime powers, where . Consider the following diagram of the first few values:
Start | Substitution | Result |
If we keep listing more values, can we find some integers , such that ?
Note: denotes the set of all positive natural numbers .
This section requires Javascript.
You are seeing this because something didn't load right. We suggest you, (a) try
refreshing the page, (b) enabling javascript if it is disabled on your browser and,
finally, (c)
loading the
non-javascript version of this page
. We're sorry about the hassle.
Wow, interesting problem! I only managed by a bit of casework - I'd be interested to know if there is a better solution.
The primes here are a bit of a distraction. The thing to focus on is that we're working ( m o d 1 0 ) , which means two important things when we're looking at powers a b :
All odd primes are congruent to either ± 1 ( m o d 4 ) ; this depends on the last two digits of p .
This drastically reduces the number of cases to check. Each decade of numbers can contain primes ending in some, none or all of the digits { 1 , 3 , 7 , 9 } . Depending on the parity of k , either those ending with 1 and 9 will be congruent to 1 ( m o d 4 ) and those ending with 3 and 7 will be congruent to − 1 ( m o d 4 ) , or vice-versa. This leaves a total of 2 4 − 1 = 1 5 choices of the possible prime endings in the decade, and 2 choices for the parity of k , for a total of 3 0 cases.
It's easy to check these cases by computer; we find that all results except 5 are possible.