Acetylation of Ferrocene

17. October 1996
Experiment #7
Acetylation of Ferrocene
Introduction
In this lab we will be utilizing the Friedel Crafts process of acetylation of ferrocene. Ferrocene is an atom of iron bounded by two aromatic rings. We will use some reagents that will cause the ferrocene to add either one acetyl group to an aromatic ring or add two acetyl groups to each of the aromatic rings. In order to determine how well this process had worked we employed: IR spectra analysis, column chromatography, and a little TLC. This experiment is relevant in today's highly industrialized world. By utilizing many of the techniques we employ in this lab, a company can synthesize new types of materials or composites that could revolutionize an industry.

Background
When we react the ferrocene with phosphoric acid and acetic anhydride, we obtain many disparate products. Not only do we get acetylferrocene, but we also get diacetylferrocene, some unreacted ferrocene reactant, and acetic acid as well. We will use thin layer chromatography (TLC), column chromatography (CG), and IR spectra analysis in order to determine the what proportions of each of these compounds will be present in the final product.
Both TLC and CG are excellent methods of measuring the presence of a given substance. Both methods turn around a compounds polarity. As one recalls, polarity is a measure of the electronegativity of a compound determined by their placement in the periodic chart. Specifically, in this lab we are talking about the difference in polarity between the atoms of oxygen and carbon. Ferrocene is relatively low to none in polarity. Acetylferrocene, because of the carbonyl functional group, is more polar than the ferrocene. Moreover, diacetylferrocene, because of the 2:1 ratio of the carbonyl groups over the acetylferrocene, is the most polar of the lot.
As stated above, both TLC and CG take advantage of polarity. Both methods have an extremely polar stationary phase; specifically, silica or alumina gel is used. Through this polar stationary phase, a mobile liquid phase is passed. Now, one can think of a polar stationary phase as a bully that waits in the high school halls for his hooligan friends. His hooligan friends, hooley's as I like to call them, always stay back to talk him; the rest of the normal student body simply keep walking and pass him. The idea here is: like-stays-with- like. Analogously, those compounds which are most similar to the stationary substrate will stay behind to "hang out". In this case, the more polar the compound is, then the more it will stay behind as the rest of the product moves forward in its liquid mobile phase. TLC works by capillary action, where the mobile phase is drawn up the TLC plate and across a polar TLC plate. CG, on the other hand, works by having gravity pull the liquid mobile phase down a polar laden column. The joyous wonder of TLC and CG, then, is that they are thus able to separate each constituent contained in the product.

Methods & Procedure
The procedure of for this lab may be found in the pre-lab note for this experiment contained in the appendix. I will only remark on the important features of the procedure. The amount of start material for this lab was ca. 10 g. The calculation for this may also be found in the pre-lab I first added acetic anhydride to ferrocene (FC) and then warmed to add in the H3PO4 catalyst. I observed a reddish-violet color to this mix of reactants. I then did a TLC and noted that the majority of the sample was not the original ferrocene start material. Please see the pre-lab for reproductions of the TLC plates used in this lab. Also see table 1.2 for Rf values.
As one can see, this crude's Rf is half that of the start material. This indicates that a reaction has definitely occurred. Next we performed an extraction on this sample with Methylene Chloride (MeCl) and Sodium Hydroxide (OH-). Please see the pre-lab for a picture of what the extraction looked like. Then we transferred the lower organic potion into another vial with a little sodium sulfate for drying. Then we transferred this to a tarred vial and dried off the MeCl in a nitrogen stream. MeCl is a great solvent because it evaporates easily (bp. ca. 48øC). Moreover, we used a nitrogen steam so that we could minimize the amount of moisture in regular air from being reintroduced into the sample. This was our second crude sample and we did a TLC on it with FC start material. See Tables 1.1 and 1.2 for amount of crude sample obtained and the Rf values. We allowed this to dry over till next weekend; we then performed a CG on this sample.
We placed this crude into the CG and then added three mobile solvents to it in order to separate the crude. We used Hexane (non-polar), Ethyl Acetate (medium polarity) and Methanol (Nice and polar). The sample flowed down the column and into separate tarred vials for each colored material. The first was bright yellow. The second was a deep reddish color. The third was a dark violet. I placed each vial into the N2 stream for concentration and then re-weighed each sample, called F1, F2, F3, respectively, again. The results are presented in table 1.1 at the end of the next section. After yet another class lab on this experiment, I tried to take 5 mg of each sample and place it into K-Br for IR spectra. However, after waiting till the next period for the IR, hardly any of the materials we present in each vial. This was quite inexplicable. I did manage to get melt-temps, but I was only able to scrape together enough of the F2 for IR spectra. See the appendix for the results of the IR spectra and see table 1.3 for the melt temp values.
I also took a TLC on each one as well; the values are presented in table 1.2 at the end of the next section.


Results & Observations
The results of this experiment are pretty straight forward and are summarized in tables 1.1 through 1.3 in this section.

Table 1.1 Weights and Measures
F1 F2 F3
DRAM Vial Weight: 4.6480 12.1914 12.2362
Vial & Mix: 4.6871 12.2177 12.2439
Amount Present in grams: 0.00391 0.00236 0.0077
in mg: 3.91 mg 2.36 mg 7.7 mg
1


The next table revolves around TLC results

Table 1.2 TLC Rf Values
First Crude Second Crude FC Mix AFC Mix DAFC Mix
Spot A: .38 .43, .84 .67 .54, .75 .61
Spot B: .75 .84 .74 .77 .36
Cospot A/B: .38, .75 .43, .84 .74 .77 .36, .88
2


The melt-temp values are:

Table 1.3 Melt-Temp Values
F1 F2 F3
Sample Melt-Temp: 195-200øC 90-94øC 75-80øC
Actual Melt-Temp: 173øC 85-86øC 130-131øC
3

The IR spectra may be found in the appendix.


Discussion
From the first crude sample's TLC that was taken, one can see that roughly half of the material present is composed of acetylferrocene and diacetylferrocene. Thus we continued along in the procedures. The results of the CG, which was performed next separated the constituent products enough that more TLC's were able top be taken. The results of these TLC Rf's tell us that our separation was pretty successful. As one can see, F1 spot A's Rf, we have at least 90% of FC in separated mixture. This is great. It means the extraction was a success. The F2 percent difference is 2.7%. This means that over 97% of this material are indeed AFC. OUTSTANDING results. This was a success for, one of the very few in this lab; thus, I am quite happy. The percent difference for the DAFC, however, is quite disappointing. Only 30 % of this mix is indeed DAFC. This is not that good of a separation. The reasons for this are explainable though. I believe it is due to inaccurate CG technique. I did not wait for all of the AFC to finish flowing out of the column before I placed the vial down to obtain the DAFC. Thus, as the TLC shows, most of this material is not DAFC. I think from comparison Rf's to the AFC TLC plates that it is probably AFC; this does then confirm that its probably the AFC. Moreover, it also could be some acetic alcohol as well. If this was a side-product formed. After all, acetic alcohol is quite polar; thus it would be one of the last products out along with the DAFC. Table 1.4 shows the calculations.

Table 1.4 Percent Different Calculations
4
F1 %D = Actual - Start Material x 100
Start Material
= 0.67 - 0.74 x 100 = 9.97%
.74
F2 %D = Actual - Start Material x 100
Start Material
= 0..75 - 0.77 x 100 = 2.65%
.77
F3 %D = Actual - Start Material x 100
Start Material
= 0.61 - 0.36 x 100 = 69.5%
.36
5

The IR spectra for F2 show us that we have mostly AFC formed. The peak at the 1700 range indicates that c=o bonds are present in this sample. AFC does indeed have the carbonyl bound present. Furthermore, since the size of the peak indicated the amounts of c=o bonds present, we would expect it to be a smaller peak than a F3 IR for DAFC. But I do not have an IR for F3 for comparison. Nevertheless, it probably has a big peak there in proportion to the F2 anyway. Hmmmmm. A large peak around the 2900 mark indicates the present of a hydrogen bonded to an aromatic ring. Our IR spectra for F2 does indeed show a peak in this range. So we conclude that this sample is acetylferrocene.

Conclusions
In sum, this lab was successful. It taught us how to correctly test the accuracy of a synthesis reaction; specifically, the acetylation of ferrocene. Our results show us the accurately synthesized and separated out the ferrocene and acetylferrocene from the reaction. The separation of the diacetylferrocene was not as successful as the extraction of the other two, but an explanation for this seems superficially valid.


Make this field active and press CTRL-V