Acho estes 2 problemas bem interessantes, e não são muito difíceis. Convido os que gostam de matemática a resolvê-los:
1) Mostre que, se r1 e r2 são racionais não negativos, então raiz(r1) + raiz(r2) é irracional se, e somente se, raiz(r1) e raiz(r2) forem racionais.
2) Mostre que, se f:R-->R for periódica e não constante, então a função g dada por g(x) = f(1/x), x<>0, não tem limite em x =0.
In English (since I have friends who are English speakers)
I find these 2 problems interesting, though not so difficult:
1) Show that, if r1 e r2 are non negative rationals, then sqrt(r1) + sqrt(r2) is rational if, and only if, sqrt(r1) and sqrt(r2) are rational.
2) Show that, if :R-->R is periodic and non constant, then the function g given by g(x) = f(1/x), x<>0 does not have a limit at x =0.
2007-08-24
05:33:37
·
3 respostas
·
perguntado por
Steiner
7
em
Ciências e Matemática
➔ Matemática
Oh, houve um engano n o texto em Português:
O certo é
1) Mostre que, se r1 e r2 são racionais não negativos, então raiz(r1) + raiz(r2) é RACIONAL se, e somente se, raiz(r1) e raiz(r2) forem racionais.
Claro.
The english text is correct
2007-08-24
05:37:13 ·
update #1
Another solution for (1)
Suppose S = sqrt(r1) + sqrt(r2) is rational and admit, by way of contradiction, that one of the numbers sqrt(r1) and sqrt(r2), say sqrt(r1), WLOG, is irrational. Then sqrt(r2) = S - sqrt(r1) => r2 = S^2 - 2S sqrt(r1) + r1. By assumption, S is rational, so that so is S^2. And since, also by assumption, r1 is rational and sqrt(r1) is irrational, it follows r2 is given by the sum of 2 rationals and 1 irrational, which implies it is irrational. But this contradicts the basic assumption that r2 is rational, proving the assertion.
2007-08-24
09:22:59 ·
update #2
Another solution for (2) (very similar to the one nealjking gave):
Let p> 0 be a period of f. Since f is not constant, there are a and b in [0, p] with f(a) <> f(b). Let the sequence x_n be defined by
x_n = 1/(n*p + a)) if n is odd
x_n = 1/(n*p + b)) if n is even
Since n*p -> oo as n --> oo, x_n --> 0.
And, in addition,
g(x_n) = f(1/x_n) = f(n*p + a) = f(a) if n is odd, and, similarly,
g(x_n) = f(1/x_n) = f(n*p + b) = f(b) if n is even.
So, g(x_n) is the sequence (f(a), f(b), f(a), f(b).....). Since f(a) <> f(b), g(x_n) doesn't converge, though x_n --> 0. Therefore, g can't have a limit at x = 0.
2007-08-24
09:33:12 ·
update #3