Saturday, December 4, 2010

Emily's Chemistry Adventures: Flourless Chocolate-Hazelnut Cake with Nutella Powder and Frangelico Caviar

I got a big box of chemicals in the mail the other day, and am looking forward to playing with them. My pain issues are making it hard for me to spend 13 hours in the kitchen making fun, elaborate dishes these days, so I thought molecular gastronomy would be a way to keep having fun in the kitchen without spending quite so much time in there. There will probably be a lot of failures to go along with any successes, and I promise to report on both. I'm easing into things by starting with two garnishes for Mini Flourless Chocolate-Hazelnut Cakes.
Today's Chemicals:
  • Tapioca Maltodextrin
  • Sodium Alginate
  • Calcium Chloride
The cake recipe came from Epicurious, and I used my miniature cheesecake pans to make adorable single-serving versions. I made one major change, which I'll mention below, and also halved the recipe so it would make 12 mini cakes. Things start with (Amish) butter and 60% cacao bittersweet chocolate gently melted over a double boiler.
Eggs, brown sugar, and Frangelico are whisked together, then the melted chocolate is added and whisked in. At this point it's time for the "hazelnut" part of the recipe. As written, you're supposed to add hazelnuts that you've ground in your food processor. I did this when I made this cake a few years ago (testing it to use as a dinner party dessert while I was in Melbourne), and found that it's just not possible to grind the hazelnuts fine enough for them not to interfere with the texture of the cake. This time I ordered some Italian hazelnut flour from L'Epicerie while I was ordering my big box of chemicals, and the texture was gorgeous.
The mini cheesecake tin is buttered and the base is wrapped in foil...
... then the batter goes in. I think I over-filled them a bit, but I (sort of) found a solution to the resulting problem...
... which was that after the requisite 28 minutes in a 350°F oven the cakes puffed up a lot. (This could also be related to the fact that I forgot to tent foil over the pan while they cooked like I was supposed to... Oops.)
I tried a sample mini cake at this time to make sure they were properly cooked, and Wow. So awesome. At least I knew there would be one success no matter how the chemistry went. No garnish needed, but these were my excuse to play with chemicals, so I forged ahead. The first garnish is one of the easier types of fun you can have with molecular gastronomy: powdered nutella.
I had a recipe for the proportions, but I ended up not following it at all (as is my way). For my first try, I used 80g of nutella with 20g of tapioca maltodextrin, buzzing the ingredients together in my mini food processor until the mixture had the consistency of soil. (The recipe called for 3 parts tapioca maltodextrin to 4 parts nutella, which is quite a different ratio.)
I dumped this onto some parchment and plated a bit in an amuse bouche glass.
It's super-awesome, because it's a light, airy powder but turns into sticky, gooey nutella in your mouth. Like magic! There was some nutella still stuck to the walls of the processor (next time I'll add the tapioca maltodextrin to the bowl before the nutella...), so I scraped that down and added a couple more big spoonfuls of tapioca maltodextrin to the bowl. After combining, I sifted through a relatively fine strainer and had a more sand-like texture.
I would love to tell you the proportions I used in this batch, but I sort of eyeballed it... which is why I'm not a chemist. Look how pretty:
Here are my two nutella powders side-by-side for comparison. I like both in different ways. Next time I'll quantify the amounts on the finer one.
The second garnish is spherified Frangelico. The ideal way to spherify alcohol is with calcium gluconolactate, but L'Epicerie was sold out when I ordered, so I had to improvise. Sodium alginate doesn't dissolve well in alcohol, but I read here about first dissolving the chemical in water, then adding the liquor, so decided to give that a try. I like the giant warning label on the calcium chloride bag. Hee.
I measured out 1.25g of calcium chloride and dissolved into 125g of bottled water, then added an additional 125g of water once it was dissolved, thus completing the "setting" bath.
Next I weighed out a gram of sodium alginate. I went off-recipe here, too.
I didn't want my Frangelico diluted as much as the recipe (which does a 50% dilution on rum...), so I started by trying to dissolve the 1g sodium alginate in 31.1g of water. When it didn't go in easily, I microwaved for 20 seconds and whisked to end up with a thick goo.
To this I added 93.9g of Frangelico and whisked to combine. I let everything settle for a while before I got to making caviar... Things were still a little more diluted-tasting than I would prefer, and I think I can remedy that once my calcium gluconolactate arrives.
I'm going to need to set up for video one of the times that I use this technique. The technique is super easy and is fun to do, but it's hard to photograph while doing it. You basically load up the Frangelico mixture into a syringe and squeeze out drops into the setting bath. After 30 to 45 seconds, the gel shell should be formed and you can transfer the spheres to a rinsing bath... and you're done. Here's my first batch sitting on my cutting board:
My second batch in a bowl:
And my third batch, after I cut off the end of the syringe to try to get larger spheres (which instead resulted in more varied sizes... I think I need a bigger syringe...).
To show you how cool the technique is, I made a large (oblong) "sphere"...
... then punctured with the tip of a knife...
Cool, yeah? On to the final plating. My over-filling of the cheesecake pan resulted in a few of the cakes being rather funny-looking.
I sliced off the top of the cake (keeping the cut parallel to the bottom of the cake), set aside the top to enjoy with coffee in the morning, and plated up the now-perfectly-cylindrical base. I first spooned over some of the nutella powder, which I kind of like the look of. (I spent all of 2 seconds thinking about plating, so these may look better next time I make this stuff...)
After the nutella powder, I spooned some Frangelico caviar onto the top of the cake.
It's not a traditionally pretty dessert, but it's kind of cool-looking, I think... and it's definitely different than your typical flourless chocolate cake plating. These cakes were velvety and delicious. They're so easy to make, and are honestly one of the best chocolate desserts I've ever made. Using hazelnut flour transformed the cake from "pretty good, but not good enough to serve to guests" to "fan-freaking-tastic." I love the nutella powder and will play with other uses for it. The caviar was a little diluted-tasting, but I think that can be remedied. All in all, a very good first experiment.


Anonymous said...

I want to play with chemicals too!

emmo said...

I'll bring some home at Christmas... Hopefully the TSA people won't steal them. =)

Anonymous said...

Thanks for your post. I’ve been thinking about writing a very comparable post over the last couple of weeks, I’ll probably keep it short and sweet and link to this instead if thats cool. impression material