Something isn't right with this story. They way institutional review boards work, there is no way that anyone would approve a study that has a risk of death, unless the study involved incredibly sick people who needed a long shot type treatment. If this is true that UNC was doing this, they risk losing their entire IRB accreditation, and no one in their right mind would risk that, as human studies typically bring the most money in the door. Also, even if its an EPA funded study, the IRB is controlled by the institution at which the research is being performed, not by the funding agency. Leave it to the Washington Times to not make this clear. The way IRBs work isn't exactly common knowledge unless you're involved with them.
Yes, if this happened as they say it did, it must have been a couple researches going off the plan. There is no way an IRB would approve this type of protocol. But in that case you could never publish your findings anyway, so you really don't gain anything. We'll have to let this story develop to find out what happened (although my guess is some lawyer is extorting the gov't/UNC and they'll get an out of court secret settlement).