How to replace all the occurences of chars like '[' , '(' etc with '\[' , '\(' using "sed" in a file?

I have a file "say file.txt" with following contents:

Capsule arr[0] in state A
rate_ul/dl=(2000000/7000000)
Capsule RBx[0] in state
...
...

using sed operator how can i replace all occurences of [ with \[, ( with \(, ] with \] and so on.

Capsule arr\[0\] in state A
rate_ul/dl=\(2000000/7000000\)
Capsule RBx\[0\] in state
...
...

Using the substitue operator in "gvim" I am able to achieve the same result. ie. if i use :1,$ s/\[/\\[/g in the vi editor in command mode I see all the [ chars replaced with \[.

However if I try to use the same substitue command in a shell script using a sed command, i am not able to achieve the same result. ie If i use the following command in a shell script I am not able to achieve the desired result:

sed "s/\[/\\[/g"  $temp_file2 > $temp_file1
where $temp_file2 conatins the lines with '[' characte开发者_如何学编程rs and $temp_file1 should contain the replaced '\[' chars

 sed 's/[][()]/\\&/g' infile > outfile

Output

$ sed 's/[][()]/\\&/g' infile
Capsule arr\[0\] in state A
rate_ul/dl=\(2000000/7000000\)
Capsule RBx\[0\] in state

gawk version

gawk '{gsub(/[()\]\[]/,"\\\\&")}1' file