Synthesis of tetraethyl orthosilicate

Some time ago I prepared some Si(OEt)₄ via silicon sulphide and ethanol. My reasoning for this was that it would avoid having to go via silicon tetrachloride, which a notoriously low-yielding reaction. I would, however, generate hydrogen sulphide, but on the scale I was working, that was not difficult to deal with.
Prompted by Eddygp, I shall write up the process here.

Extraction of silicon from sand
Sand (10 g) was mixed thoroughly with magnesium powder (8 g). The mixture was piled onto a steel plate and ignited with a blowtorch. I relied on the magnesium powder as a ignition initiator and reducing agent. When the resulting friable mass had cooled down, it was digested with hydrochloric acid (50 mL, 31%) and allowed to react overnight. The black precipitate was filtered off, washed with water, and air-dried.

Synthesis of silicon sulphide
My first idea was to attempt union of the elements. Silicon (3 g), and sulphur (6 g) were mixed and heated to a bright red heat in a steel crucible with a blowtorch. However, I could not achieve the temperatures required to initiate a reaction. I hypothesised that my volume was too large for the heating method I was using. So I tried the same experiment in a test tube, but again to no avail. I researched further into the reaction, and discovered to my chagrain that I would have to have heated the reaction mixture to about 1500°C to achieve a reaction.
My next idea was to use a method I came across in the video on silicon extraction by hkparker of youtube. He mixed sand, aluminium powder, and sulphur. The reaction between sand and aluminium powder is too endothermic to be a viable method of obtaining silicon, but the reaction between aluminium and sulphur is tremendously exothermic. Hence, these two reactions are combined to achieve sufficient heating for the desired reaction (6Al + 3SiO2 -> 2Al2O3 + Si) to take place.
So, I tried this principle. Silicon (3 g), aluminium (3 g), and sulphur (12 g) were mixed thoroughly and heated in a steel crucible (mistake). Upon ignition of the mixture, it reached a glowing white-heat (no way of measuring temperature) and melted through the crucible with great ease. This was allowed to cool, and the whole mixture of reaction product and crucible was put into azeotropic ethanol.

Synthesis of tetraethyl orthosilicate also known as tetraethoxysilane
The reaction mixture was allowed to sit and evolve gases outside overnight, and the following day the remains of the crucible was removed and the mixture filtered to remove unreacted particulate. The ethanol was distilled off, leaving a high boiling residue that I considered to be my final product. It was a rich orange-brown liquid that produced a white precipitate when a little was added to water. The resulting aqueous solution reduced acidified dichromate.
On the basis that the liquid was high boiling and did not distil readily, and was hydrolysed by water to give a material that reduced acidified dichromate, I concluded that I formed Si(OEt)₄. I admit that there may well be Al(OEt)₃ contamination also present. The red colouration I put down to Fe(OEt)₃. Interestingly, over the course of the following days and weeks, this precipitated out leaving a pale yellow liquid with the same behaviour as documented above.

High boiling residue:
Final product:
Reaction with water and dichromate:
A little time later:

Unfortunately, I do not have any pictures of the product much much later, when all the red-orange material had precipitated out.