back to main page

Why a lava cake?

Though I planned to make a chocolate-lime cake this week, I’ve decided to postpone it. After thinking a little bit about my choice this week, I decided that I wanted to make a frosting-free cake. None of my guests particularly like frosting, and it’s about time to change the pace from the “decorated layer cake” mold I’ve stuck myself into—especially because I won’t need a cake to decorate for the Wilton class this Saturday. (We’re only making icing flowers.)

After a little more thinking, I settled on a “molten” (or “lava”) cake. They’re so hot right now (or at least they were hot a year and a half ago), and I’ve been terribly late to the game. Before, I figured I didn’t want to get into that fad-wagon, but now that the fervor has passed, it’s time to move in on it. (Plus, I’m still convinced Andrew wants to eat another one after our great dinner at La Bella Trattoria for his birthday.) I thought, “Lime would pair wonderfully with the intense flavor of a flourless cake, and I could still mix up some lime whipped cream or some hot fudge for a topping if necessary.”

But this choice gave me fewer options for the lime additions, so I decided to wait on the lime and go molten:

Recipes

1. lava cupcakes use a muffin tin and Grand Marnier (so how about replacing w/Chambord?) #individual lava cakes (a basic recipe)

#by Pam Anderson, the one responsible for the awesome scones

Experimentation, round 1

After the fun experimental process involved with the Rose Pistachio cake, I decided to recreate it. I chose the Grand Marnier recipe because 1) it’s made in muffin tins, therefore expediting the experimental process, and 2) the recipe includes liqueur in it, so it has built-in room for flavor additions. Instead of just willy-nilly adding flavors, then, I’ll just substitute the liqueur for an extract or other liqueur.

In this case, I have 12 potential flavors. Yesterday, I received a package of eight all-natural essences/extracts: caramel, strawberry, rose, rum, black walnut, lime, and lavender. That’s eight. We have a pretty good selection of liqueurs, so I can use Chambord, Starbucks, Frangelico…and whiskey (for Andrew). That’s twelve.

As it turned out, I only had enough batter for 11 cupcakes, and I couldn’t get the Frangelico opened, so that’s what was left out. So little batter resulted, in fact, that I only had very small cupcakes for some, and I worried about having a small enough amount of extract balance the small amount of batter. Each cupcake, I calculated, would receive 1/2 tsp of the liquid, so I planned accordingly, yet with the little amount of batter, I had to kind of wing it as for the amounts.

Seven minutes at 400F cooked them perfectly, with little molten centers in most of them:

They came out, well, “indifferent.” They popped out of the muffin tin just fine, and the consistency and flavor of the cake itself was fine. The extracts, however, were none so impressive!

Two of them exploded upon exiting the pans, prompting looks of concern:

Results

1. The caramel tasted way too artificial.
2. The black walnut tasted okay (I’ll use it again)
3. The strawberry tasted fruity, but not particularly like strawberry
4. The lavender cupcake smelled and tasted like gin (phew!)
5. The rose made Andrew gag and spit it out (we think he’s mildly allergic)
6. The rum extract was absolutely disgusting (it was all you could taste)
7. The Starbucks liqueur completely disappeared
8. The lime cake tasted actually very good, quite jazzy and fun
9. The cinnamon didn’t impress Andrew, but it was subtle and interesting
10. The Chambord baked off: we couldn’t taste it at all!
11. The whiskey was quite interesting and yummy

Overall, then, the black walnut, strawberry, lime, and whiskey worked out pretty well. I’m glad the lime-chocolate combo worked out, as I’ll be working with that combination again very soon, and the whiskey was a surprise hit. Overall, we learned that artificial tasting extracts don’t taste better after baking, and that you have to use a lot of liqueur to ensure that it survives the baking. Also, we learned that the muffin tins made the bottoms of the cakes a little too brown and overdone, and that cooking the cakes straight from the fridge did ensure perfect molten centers.

Andrew in particular was not a fan of many of the extracts, so he came up with a great substitute: dressing a “plain” lava cake with creme anglaise (a simple custard sauce) and raspberry coulis. Beranbaum has a recipe for raspberry sauce that she claims is the best she’s ever had (not a modest one, that cook), and I’d like to try it out.

One problem with this recipe was the flour content: it only contained 1/4 cup (4 tablespoons), but it still tasted very floury. Neither of us liked the fact that it still tasted like regular ol’ cake, or, alas, like a good-quality brownie. As a result, I can’t use Pam Anderson’s recipe (she uses the same amount of flour next to 25% more chocolate—not a drastic enough change for me). We’re down to the Epicurious recipe, then! We’ve got a winner.

More Pregaming: nitty-gritty research & details

Now, I’m looking for cooking tips about bakeware. The muffin tins did work great, except that the outsides looked a little too dark. I decided therefore to explore my options and find out what other people generally do. Overall, the consensus is (drumroll) ceramic ramekins. Now, we have tiny 4 oz ramekins (for prepware, storing chopped herbs and other small ingredients), a four-pack of classic 5 oz Le Creuset? ramekins, and flattish 10 oz Martha Stewart ramekins (for the larger ingredients)—but no standard 6 oz ramekins. While the smaller ones will be too small to satisfy any sweet tooth, the larger, flattish ones would make the molten center impossible.

The next option, my mini-cake pans, are too large. Four inches in diameter, each produce two servings’ worth of cake, and because the molten cake is generally very rich, that’s a no-go.

What am I left with? My Pyrex custard cups. They’re 6 oz and ovenproof…but they’re clear. I toyed with the idea of placing them in a water bath (like traditional custards), but we don’t want a mousse texture, so I think I’ll just take a chance on it. A small body of lava-cake-bakers have used these cups, so I feel pretty confident about the choice. Also, we decided to perform another test-run before the Tuesday cake n cocktails night. (Andrew is always quite concerned with presenting our guests only the very best we can offer, so he’s pretty compliant about my constant testing.)

The next job: scour recipe forums for the small pieces of advice you don’t find in the recipes.

Tips I found:

1. Use high-quality chocolate (Sharffen-Berger, Callebaut, Perugina, etc; astonishingly, we found 1 forlorn container of dark Callebaut on sale at Giant groceries).
2. If the recipe uses only one kind of chocolate, add 1 tsp instant coffee crystals and/or 1 tsp unsweetened cocoa to boost the chocolate flavor
3. Grease dishes with butter, then coat with sugar and cocoa (for easy removal; although I will use shortening instead, to avoid over-browning; sugar and cocoa mixture should be 1 1/2 tsp for each ramekin!)
4. Don’t overfill the custard cups (at the most, 3/4 full)
5. Refrigerate cups for 20 minutes before placing in oven (most recommend putting them in at room temp, but this method results in over-runny centers)
6. 400F oven is best (hotter, but shorter baking time, ~11 minutes)
7. Don’t overcook: only wait until the top has cracked and bubbles
8. Let them sit in ramekins, cooling, for two minutes until you invert them to remove, and make sure to run a knife between cake and ramekin before inverting (but they should still be warm when you invert); also, invert directly onto dessert plate)

Second Round of Experiments

For the second round of experiments, we made a few changes. First of all, we used actual blocks of chocolate, rather than just gourmet chocolate chips:

The silkiness of the chocolate melting in butter can’t be underestimated at this point. Oh, beautiful! (Almost as much as Andrew’s demi-glace the other night.)

The other changes involved recipe changes (which you already know about), and the experiment was centered around the cups themselves. Some were prepared with Crisco, some with butter, some also with sugar, and some also with cocoa. Some cakes would have espresso added, and others would have extra cocoa:

This new recipe allowed me to view a few really neat chemical changes in materials due to folding, emulsions, and whipping. For example, the 8 minutes of beating eggs and extra yolks showed a rainbow of yellows, starting with the deep yellow-orange of yolks, then melding into a buttercup yellow, and finally into a light, creamy, butter-yellow.

At the beginning of beating:

Five minutes into the beating:

And at the end of the eight minutes:

I love how the change shows up so well on the digital camera! The other massive change involved the next step, folding the chocolate into the egg mixture (I know that you’re not supposed to fold in the heavier item into the lighter, but I wanted to follow the recipe, at least this time). I knew the 8 minutes of beating had done the trick when the beaten eggs actually sizzled a little when the chocolate was added!

This folding was tricky. The substances truly avoided each other, slipping out of one another’s grasp for a full three or four minutes (astonishingly long for a process that normally takes under a minute).

After a minute:

The mixture begins to darken:

Inexplicably, the ingredients jump in and have a tea party together at the end of a couple of minutes:

This mixture was so thick and shiny that I could have sworn it was ganache. Andrew kept licking the bowl clean; it must have been the Callebaut chocolate that made the difference, we theorized.

After baking, we had seven sample cakes stretched out in front of us:

From the taste test, I learned some unexpected lessons. First of all, even though cooks warn about using butter instead of shortening, the Crisco did not help the lava cakes at all. The cups were only in the oven for a little over ten minutes, so the butter-lined cups didn’t even threaten to scorch or over-darken. Furthermore, the butter-lined cups actually released from the cups more easily than the shortening ones! In general, the butter-lined cakes tasted better, while the Crisco-lined ones left an unpleasant aftertaste.

With one exception: the sugar-and-cocoa dusted, Crisco-treated cups tasted great and lifted out of the cups much better than the butter ones. With the Crisco flavor knocked into oblivion by the cocoa, we could tell that the butter-lined cakes did not have the in-your-face chocolate flavor that the Crisco-ones did. In effect, the butter masked the chocolate. The Crisco didn’t, but it had to be tamed by the addition of a cocoa dusting.

The additions to the batter (extra cocoa and espresso) were unnecessary. The espresso did “pump up” the cocoa flavor, but it left a funny smell. We could effectively amp up the chocolate flavor by dusting the cups with cocoa, so there was no need to add the espresso after all. Meanwhile, the extra cocoa in the batter didn’t do very much for the cakes; they didn’t have enough sugar to make the extra cocoa really jump out as a definitively chocolatey flavor.

The winner, then, was the Crisco-lined, sugar-and-cocoa dusted cup! It did turn out, though, that you do need the apparently excessive 1-1/2 tsp of sugar-cocoa mixture to dust each cup with. Without the extra layer of dust on the bottom of the cup, the bottom of the cake stuck to the cup when inverted.

On the night of the dinner party, we prepared creme anglaise and raspberry puree to plate with the cakes: yummy! Despite an emergency maneuver (placing the cakes back in the oven after inverting one unfortunately underdone lava cake—more lava than cake), they were a success.

What I Learned

1. Don’t be afraid of beating eggs on HIGH, not medium speed
2. Crisco leaves a bad aftertaste unless masked by some kind of dusting
3. Butter won’t scorch on a small pan left in the oven for a short time, nor make a cake stick to a small pan
4. Timing lava cakes is a nightmare: one minute on either side can ruin the effect

back to main page