Some initial observations from dis-assembly. .The silicon top seal seems to be in good shape. .There are no visible irregularities around the outer circumference or at the air tube hole. .The piercing from the syringe looks clean and .has a sort of arch shape that appears to close tightly. .The unit seemed to have a good seal since after the silicon seal was removed the liquid remaining in it leaked out of the battery end.
Examination of the barrel, or outer shell, revealed many stress cracks, some of which appear to go through the wall entirely. .There are four areas that are longitudinal with the longest being almost .75" long and the shortest being .625" long. .They are located at about 90 degree intervals around the barrel which would indicate, to me, that they are a result of tension being set too high on machinery that handled the barrel during manufacture. .There is another area of stress fractures at the base where the barrel mounts to the atomizer assembly. .Again, some of the fracture line appear to go completely through the wall of the outer barrel. .The wicks are set 90 degrees from the wick ports, which I think, is normal. .The coil, itself, is clean with no apparent residue from heating.
Based on what I have seen with this one so far, I think the problem came once the liquid got below the level of the array of stress fractures in the barrel. .They are almost uniform in where they start, which is about .5" from the top of the barrel.
If you experience the problem of leaks 1/2 way through a CE3 full of liquid, examine it by holding it horizontal to a source of light and rotate it. .Stress cracks will show up as whiteish lines in the plastic. .I think that is the culprit.