Making a pastel de nata isn’t rocket science – or is it?
I’ve been making endless tarts, experimenting with my own recipes. They’ll all delicious. The trouble is they’re not natas…
Genuine natas are made by Pastéis de Belém in Lisbon Portugal following an ancient secret recipe. So we can’t get any help from there. I Love Nata in Seven Dials bakes their natas on the premises but the filling is imported from Portugal. I asked for a recipe – they smile politely, but say they don’t have it. Canela, the Portuguese cafe and restaurant did once make their own nata, but even they had to admit it was easier to have their tarts imported by a specialist Portuguese supplier. I have the telephone number of the company, but can’t imagine them giving me the keys to nata fortune.
Something fishy is going on here. Why is it so difficult to replicate the tarts? I think there is industrial/chemical culinary science going on here (if not rocket science!). Unfortunately, for us home cooks the most sophisticated science we can use to unlock the secrets of this kind of recipe is trial and error. Here are some of my trials – and errors…
A nata is not an English custard tart
So if you follow one of these types of recipes the result will be very nice but not a nata. What’s different? The custard – it’s custard, but not as we know it.
What’s in the custard?
Natas have a smooth emulsified consistency, they are paste-like but still wet and soft. They don’t have that delicate wobbly, jelly-like consistency of baked custard. Indeed, they have a kind of set reminiscent of panna cotta (but don’t try cooking your cream for a nata – it doesn’t work). Nata also remind me of the Greek custard pastry galaktoboureko – the key ingredient of which is semolina flour. The other taste I get is the old-fashioned vanilla slice – basically a creme pat between sheets of pastry. The creme pat has corn flour or flour as a main ingredient and also includes butter. Several online sources use milk and heated sugar syrup – this might be similar to making a sabayon for ice cream but the hot sugar is not added to the eggs so I do not see the benefit. Other ingredients could include milk powder, egg white powder, or UHT milk or cream all of which are used to ‘stabilise’ egg mixtures and corn syrup which gives bakes a moist consistency. I have tried corn flour, normal flour and semolina. I settle on semolina as it gives a better taste and texture although it makes a slightly cakey top to the tart. I also use butter and corn syrup.
Would Pastéis de Belém really use this much cream and eggs?
Some recipes call for cream and huge amounts of egg yolks. But I wonder how it is possible to sustain such quantities of precious raw materials for a mass produced product, let alone justify the reckless waste of egg whites. My recipe incorporates a slightly more balanced use of eggs.
What about the heat?
We understand the tarts are cooked at 250C which is a little beyond what domestic ovens can achieve – or even what some non-stick muffin tins are safe to bake at. The tarts cook in 10-15 minutes at this temperature will develop the characteristic burnt tops. However, my pastry expands and pushes the contents over the edges of the pastry case. So I have baked the pastry blind. I have also tried a lower heat method, but the high heat seems to cook the custard to a finer texture. I have suggested both methods in the recipe.
And the cinnamon?
The pastels are traditionally served with icing sugar and cinnamon sprinkled over the tarts. I have tried putting cinnamon in the tart but it doesn’t give the right taste. It’s best to add the cinnamon after.
The perfect nata?
I have no doubt I will eventually be able to replicate a genuine nata – but not without a hundred more eggs and several gallons of cream…
So it is time to stop. My latest version has a good consistency and taste and is sufficiently different to a custard tart for me to confidently say I have made a damn good nata.