After it was brought to a boil, we removed the condenser and added the two solutions we had made into the flask. At this point, it turned a dark green color.
We let the solution cool to room temperature and we saw that something had precipitated in the solution. We vacuum-filtered it and got a green-gold precipitate.
We let the precipitate dry overnight and then we extracted this solid with 20 ml of methylene chloride, and the methylene chloride turned yellow.
We put a boiling stone in and boiled off the solvent until there was very little solvent left in the beaker. We cooled this in an ice bath, and crystals were supposed to form at this point, but we didn't really see anything happening. There was some kind of precipitate after a while, so we added some hexane and boiled off the solution again. We got some yellow-brown solid at this point.
We took the melting point of this solid, which was about 190 degrees Celsius. However, we did not have any reference data for this compound, so we couldn't really tell if this would help us to identify that we had made the product that we wanted to.
For the part 2 of the Wild Card, we placed triphenylphosphine in a round-bottom flask with 12 ml of absolute ethanol and stirred it until it dissolved. We attached a water condenser to the flask and heated the solution until it boiled. While we were waiting for the solution to boil, we prepared a solution of 50 mg of ruthenium (III) chloride hydrate in 4 ml of ethanol and a solution of 80 mg of Diazald in 4 ml of ethanol. We added the ruthenium solution to the flask first, at which time the solution was a dark brown.
We added triethylamine to the flask via pipet until the solution turned a dark reddish-purple. At this point, we added the Diazald to the solution and it turned a dark yellow. There was a precipitate that we vacuum-filtered and it turned out to be a reddish, coppery precipitate.
The manual, however, said that this was supposed to be a gray precipitate! We took the melting point of the solid and got 140 degrees Celsius. We have to do a KBr pellet IR on each of these, but we have yet to do this. Again, we have no reference for the IR data, so we have no way to determine whether we have made the products we were supposed to. Hopefully we can get the IR to work for us so we can see the NO stretches and compare them.
No comments:
Post a Comment