Narrative:

Departed with bleeds off and APU bleed feeding left pack due to desire to maximize stopping margin on a slick runway. Moderate snow falling and aircraft was deiced with type iv fluid. All procedures were followed slowly and carefully. Just past V1 (109 KIAS); the cockpit began to fill with massive amounts of smoke. I rotated at vr and called for the gear. Within a second or two; it was becoming very difficult to see the flight instruments. I engaged the autopilot at 400 ft AGL; removed my headset; and donned my oxygen mask as did my first officer. We established communication and I told him to check with the flight attendants to see if they had smoke in the cabin and where it might be coming from. About then; I began to suspect that the smoke was coming from deice fluid being ingested into the APU compressor. The smoke had become extremely thick and I was planning to divert; but out of about 2;000 ft MSL it began to dissipate. I removed my O2 mask and could then easily confirm that the smell was deice fluid. We were level at 3;000 ft MSL with the flaps still at takeoff setting on the tower-assigned heading of 250 at V2. I twisted the speed bug to the flaps up maneuvering speed and cleaned the airplane up on schedule. After a short period while we were climbing; we realized we had not yet switched the bleeds. We did so and the rest of the flight was normal. I am quite sure that the fuselage was deiced with type iv fluid and the fluid sheered off just past V1 and entered the APU inlet. Make certain that type iv fluid is not used on the fuselage if a bleeds off departure is planned.

Related reports:

Google
 

Original NASA ASRS Text

Title: B737 flight crew experiences a smoke filled cockpit just at rotation caused by deice fluid being ingested into the APU during a bleeds off takeoff. Oxygen masks are donned and the autopilot engaged at 400 FT AGL and as the aircraft begins to level at 2;000 FT on assigned heading at V2 the smoke begins to clear. After clean up and bleed reconfiguration; the flight continues to destination.

Narrative: Departed with bleeds off and APU bleed feeding left pack due to desire to maximize stopping margin on a slick runway. Moderate snow falling and aircraft was deiced with Type IV fluid. All procedures were followed slowly and carefully. Just past V1 (109 KIAS); the cockpit began to fill with massive amounts of smoke. I rotated at VR and called for the gear. Within a second or two; it was becoming very difficult to see the flight instruments. I engaged the autopilot at 400 FT AGL; removed my headset; and donned my oxygen mask as did my First Officer. We established communication and I told him to check with the flight attendants to see if they had smoke in the cabin and where it might be coming from. About then; I began to suspect that the smoke was coming from deice fluid being ingested into the APU compressor. The smoke had become extremely thick and I was planning to divert; but out of about 2;000 FT MSL it began to dissipate. I removed my O2 mask and could then easily confirm that the smell was deice fluid. We were level at 3;000 FT MSL with the flaps still at takeoff setting on the Tower-assigned heading of 250 at V2. I twisted the Speed bug to the flaps up maneuvering speed and cleaned the airplane up on schedule. After a short period while we were climbing; we realized we had not yet switched the bleeds. We did so and the rest of the flight was normal. I am quite sure that the fuselage was deiced with Type IV fluid and the fluid sheered off just past V1 and entered the APU inlet. Make certain that Type IV fluid is not used on the fuselage if a bleeds off departure is planned.

Data retrieved from NASA's ASRS site as of July 2013 and automatically converted to unabbreviated mixed upper/lowercase text. This report is for informational purposes with no guarantee of accuracy. See NASA's ASRS site for official report.