Hi,
    Your second step should have an additional bake for image reversal after
the first, masked exposure.  You might be leaving some resist in the areas
that you wish to have Cr which causes the Cr to crack as they outgas in the
vaccum and expand/contract under heat loads.  The proper recipe for that
liftoff resist process can be found online (google 5214) or on the AZ
Clariant data sheet.

-Lou


>From: xiaodong wang 
>Reply-To: General MEMS discussion 
>To: [email protected]
>Subject: [mems-talk] Help: Cr cracked after lift-off
>Date: Mon, 17 Oct 2005 14:06:31 -0700 (PDT)
>
>Hi, all,
>
>I am working on a low-k material related project. The
>following is my process flow chart. Cr was used for
>etching mask for low-k material and BCB. But after
>lift-off, Cr was cracked.
>
>1. low-k substrate
>
>2. PR spin and align (AZ5214E, 4000rpm/1Krps/45s, bake
>90 degree C for 1 min, expose 3 sec, followed by flood
>expose for 200s, develop in 300 MIF for 1 min)
>
>3. E-beam evaporation of 120 nm thick Cr
>
>4. lift-off, (find Cr cracked into pieces).
>
>5. After lift-off, low-k material etch (Cr is designed
>as etching mask)
>
>6. BCB is spinned on Cr film and cured at 250 degree C
>
>7. BCB etching  (Cr is designed as etching mask)
>
>
>Anyone can check my flow chart and give me some
>suggestions.  Thanks