Coldfusion doesn't recognize japanese character unicode when searching with regular expression

Question

When try to search japanese characters with regular expression in Coldfusion, I got this code.

<cfset test1 = REFind("[\u3040-\u309F]", "あ")> (1)
<cfset test2 = REFind("[\u61]", "a")>           (2)
<cfdump var="#test#">

With (1) I got the result 0 that indicates there is no match. But actually, the above unicode matches all hiragana characters including the above char あ. It should return 1.

With (2) I got 1 that indicates it matched.

I also tried with all ANSCII chars, it works but japanese chars doesn't.

How should I fix this?