Seely English Toffee

(Another delightful and delicious delicacy extracted from Ollie's Illustrated Gourmet Cookbook)

(Updated November 26, 2016)

The recipe has been updated to show quantities of ingredients which will not be in significant excess at the end of the procedure.

1 lb sugar
3-1/2 quarter pound cubes of regular salted butter
12.5 ounces finely ground walnuts
4.5 ounces finely ground almonds
16 ounces milk chocolate
6.5 ounces unflavored malted milk
1 tsp vanilla flavoring
1/4 cup water
Heavy duty (wide) aluminum foil and chopping block large enough to hold one square piece.

Mix sugar, butter, vanilla and water in large pan. This recipe has been seen to work with as little as 1/2 lb butter (2 cubes) and as much as 1 lb (4 cubes). Using a pound of butter gives it a slightly softer "crunch" and it certainly pours better but some of the melted butter may not be absorbed completely in the sugar.

Heat over fairly high flame, stirring constantly, to beginning of hard-crack temperature range (around 305oF). High flame works well to avoid phase separation that will occur between sugar and butter if heating process is too slow, but when one approaches hard-crack temperature range, the peril of a phase separation has been passed, and the flame can be reduced to avoid burning the mixture. Once into hard-crack temperature range (295o - 315o F) one cup of packed ground walnuts is added to the mixture and stirred in. If the cup of walnuts can be preheated for 2 minutes in the microwave oven when you have reached the soft crack stage, you avoid a large drop in temperature and possibly a phase separation. Note in the image above right that the thermometer had passed into the hard crack range,
but after having added the nuts, even though they had been heated in the microwave oven, there was a slight drop (left) down to around 290o between soft crack and hard crack zones. Slowly heat the mixture, constantly stirring, until color is a golden brown. The toffee will harden at any point in the hard-crack temperature range, but some people like to heat it until the temperature moves into the caramelization temperature range. Be careful as the whole batch may turn into goo and you'll have to eat it with a spoon. There is a subtle change in taste from the mid-hard-crack point to the beginning of the caramelization point. Some people like the "barely-burnt" taste one gets by going into the caramelization zone, others do not. To be safe the first time around one ought to stop the process in the mid-hard-crack zone. In 2004 I made two batches by turning down the flame and slowly easing up to the top temperature of hard crack, around 315o F. It hardened beautifully, but to tell you the truth, I don't like the "barely-burnt" taste. It is sadly below my best batches and I'm a little embarrassed by it. My advice: pour the batch in the middle of the hard-crack zone at around 305o F and you'll be delighted with the results.

If a phase separation occurs, don't worry, just turn down the flame a bit and keep stirring gently, and be careful not to let the hot butter splash out onto your hands. The butter and sugar phases will go back together again as the temperature increases. Sometimes a phase separation occurs while in the hard crack zone. If that happens, turn the flame down a bit and keep stirring until the two phases go back together. Even if you enter the bottom of the caramel zone, the batch will usually harden. To keep the batch from being too oily, it is good to get all of the oil back into the sugar. What one sees is that when the oil has separated from the sugar the mix tends not to stick to the walls of the pan any longer, but the higher temperature of the walls soon cause oil absorption to occur at the interface and the mix begins to stick to the wall again even though the rest of the mass has a layer of separated oil sloshing around. Don't panic, just keep stirring and a few degrees higher the oil will go back in.

Here's a caveat for you. Many candy recipes work for one person but fail for another. The failure may be the result of the use of different sources of ingredients or different procedures. My daughter complained recently that one of her batches never hardened after having made it successfully several years in a row. Was it the way she stirred, the brand of butter or sugar she used or the rate at which she heated the mixture? We don't know. Her mixture certainly passed into the hard crack range. At the risk of beating a subject to death, here's a table showing temperature, time increments (delta t) and total time for a recent successful batch. Note that for this case, the total time was around 16 minutes from first heating to pouring the mixture on the aluminum foil (see below).

Fahrenheit Celsius delta t


Total t


210 99 0:00 0:00
215 102 0:20 0:20
220 105 0:13 0:33
225 107
230 110 1:03 1:36
235 113
240 116 2:39 4:15
245 119
250 121 2:04 6:19
255 124
260 127 1:43 8:02
265 129

soft crack

132 1:28 9:30
275 135
280 138 1:37 11:07
285 141
290 143 1:22 12:29
295 146

hard crack

149 1:12 13:41

add nuts

152 0:53 14:34
310 154 0:40 15:14



157 1:00 16:14

There doesn't seem to be any secret here and the whole process is over in a little more than a quarter of an hour. Still, a batch can refuse to harden after being poured onto the aluminum foil, or it can spread out on the aluminum foil by itself or be sufficiently viscous to require a knife to spread it to a desired thickness, and harden beautifully.

Pour mixture onto large piece of aluminum foil on flat surface (large foil, not the sandwich stuff). Spread with table knife to thickness of about 3/8". Be careful not to get a good skin burn by touching the top of the candy with your knuckles. Batch will harden within 15 minutes. If more than one batch is to be made, the first batch, while still hot, can be slipped off the chopping block onto another slab and then slipped onto another flat surface, like a dining room table (with padding - don't put it on a finished top while it is still hot), but this takes a little practice. The next step can be done any time after it hardens but it is best not to leave the slab overnight, as its low temperature in the morning will cause it to crack when the melted chocolate is spread over it.

Stand up solid slab on chopping block and peel off aluminum foil for next step. Lay the slab down on the separated aluminum foil.

Melt over double boiler a pound of milk chocolate. For American taste, Hershey's works fine. My French friends love toffee with Hershey's chocolate but to the European taste, Hershey's tastes a little too fermented in comparison with Nestle's.* While melting the chocolate, mix 1 cup of unflavored malted milk with 1 cup ground walnuts and 1 cup ground almonds. Thoroughly mix nuts and malted milk. This amount is almost precisely what you'll need for one batch after you've developed some experience making it. If you make more than you need, the rest of it can be used on a second batch.

Spread half of melted chocolate on one side of hardened candy. While the milk chocolate is still gooey, sprinkle nut-malted milk mixture on surface and press in. Turn batch over and repeat the process on other side. If it cracks while turning it over, just fit the pieces together and continue. Cover with another piece of foil and allow to stand overnight. In the morning, break into small chunks.

It turns out that the very special taste of this English Toffee is connected to the candy + malted milk + milk chocolate. It tastes good in any case, but the malted milk curiously imparts an added effect which one doesn't fully appreciate without tasting side-by-side a piece of the hardened candy only and a piece of the finished candy with malted milk, nuts and milk chocolate.

Don't eat it all at one sitting!

*Legend has it that in the early years of the Hershey company a batch of milk chocolate was fermented a little too long, but packaged and sold anyway. However apocryphal that story may be, to every American kid the word Hershey is synonymous with chocolate and nothing tastes quite as good. If you would like to make a taste test on this point, taste first a small piece of Nestle's, Cadbury's, Ghirardelli's or Guitard's milk chocolate. Then taste a small piece of Hershey's milk chocolate. The additional fermentation becomes obvious. One is left with the impression that for the first time in your life you'll be of the opinion that the Hershey's is a little "off" on the one hand, but "real" chocolate on the other (if you grew up in the U.S.). Everyone ought to do this taste test at least once in a lifetime.

  • Return to other related Web pages
    Department of Chemistry
    CSUDH Home Page
    Oliver's Professional Home Page