PerfectCremeBrulee

Slow,gentle heatisthekeytoperfectcustard,soforthebestcremebrulee, usechilledcream,aprotectivewaterbath,andalowoventemperature.

yobjectivewassimple-findtheper­ fectrecipeforclasiccremebrQ!ee.My standardswereh1gh-I wa·nted acus­

tardthatwaslight,firm,smooth,creamy,sweet, fragrant, andslightlyeggy, with abrown sugar crust that was bothdelicateandcrisp. And, of course, making theprocess easy andquick was alsoaconsideration. \

AsIattempted toreach thiselusive goal,try­ ingsome thirty-six variationsalong the way,I foundthattheprocess wasoneofexclusion,not inclusion.Thefewest ingredients,fewest steps, andsimplest cooki ngtechniquesdelivered the bestresults.

The Custard

Probablythebiggestchallengetomakingcreme brilleeisgettingthetextureofthecustardright.In consultingdozensofrecipes,Ifoundasurprising numberofoptionsforthecustardingredients,in-

c<:::::.' BYGENEFREELAND· ·

eludingvariationsontheeggs (eitheryolksonly orwholeeggs), thesugar(white, brown,ornone atall),theflavorings(vanilla,rum,kirsch,various liqueurs, instantespresso, cinnamon,andgrated nutmeg),andmostimportantly,thecream(heavy, whipping,orhalf-and-half).Further variations couldbefoundinthecookingtechniques,suchas thetemperatureof thecream (from boilingto chilled) andwhether tocook thecustard onthe stoveorintheoven.

Iexperimentedwithevery possible variation, butfoundthatthemostcrucial werethetypeof cream,the cooking timeandtemperature,and wherethecustardwas-cooked.(Theresultsof my experimentswith otheringredientsarelisted in "SearchingforPerfectCreme Bn1lee,"page7.)

Istartedwithasimple,traditionalcremebn1lee recipethatcallsfortwocupsofheavycreamtobe boiledforoneminute,beatenintofoureggyolks, returned tothefireoverlow flame(inadouble

boilerifdesired),thenstirred untilnearlyboiling. Themixtureisthenpoured intoagreasedbaking dish,chilled,covered with athinlayerofbrown sugar,caramelizedunderthe broiler,chilled again,andserved.

Ibegan by making separateversionsofthis recipeusingallthreetypesofcream,andcooking themintheoven,ontopofthestoveinapan,and ontopofthestoveinadouble boiler.

Thecustardmade withheavycream,which containsbetween36and40percentfat,wasway toorich;half-and-half,withbetween 10'/2and18 percentfat, made awaterycustard.Whipping cream (sometimes called light whippingcream), whichisbetween 30and36percentfat,gavethe custard thesmooth, sweet,balanced flavor and textureIwanted.

After thisfirst set of tests, Ialsodismissed

cookingthecustardonthestovetopinasaucepan, sincetheresults weresopoor.Thedoubleboiler, wasnotmuchbetter, but Idecided totrysome\ variationsbeforegivingupontrusmoreforgiving method.

Ialsodecidedto callShirleyCorriher,cooking

nmysearchfortheperfectcremebrillee,Itrieddozensofpossible variationswiththebasicin­

gredientsofcream,eggs,andsugar.Thefollowinglisthighlightssomeofmyfailuresalongthe roadtosuccessfulcremebrillee.

teacherandresidentfoodscienceadvisorfor Cook'sIllustrated,todiscuss theresults ofmy tests.Corriherstartedwithsomebasiccustardsci­ ence.She explainedthat when egg yolks are

EXPERIMENT

Heavycream usedincustard Half-and-halfusedincustard Wholeeggsusedincustard

Flavoringssuchasvanilla,cinnamon,nutmeg, andinstantespresso

Caramelized white sugar usedincustard

Brownsugar usedincustard

Saltadded tothecustard

Cornstarchaddedtocustardcookedin doubleboiler

Yo-inchbrownsugar topping

Powderedsugarused todustbutteredramekins

CONCLUSION

Toorichforitsowngood

Weakflavorandwaterytexture

Toodenseandfirm

Extractsandspicesdetractfromthesweet creamandeggflavors

Flavorisburnedandtextureisgrainy Textureisgrainy, flavortoosweet Oddandoutofplace

Mayhelppreventcurdling,butleaves behindagrainytextureandmakes custardverydense

Formsthickbarrier thatistoodifficultto penetrate

Evensmallamountsmakethecustardtoo sweet

heated,thebondsthat holdtogether theproteins

intheyolksbegintobreak.Theproteinsthenun­ windwiththeirbondsstickingout,runintoother unwound proteins,and bind together toform a three-dimensionalmesh.This is what causes a custardto thicken.Whenacustardreaches180 degrees,theproteinsbondtogethersoextensively thattheyformclumpsandtheeggscurdle-inef­ fect,theybecomescrambledeggs.

Becauseofthisdynamic,thespeedwithwhich

youheatthecustard mixturesisveryimportant. "Ifthe eggs are heatedquickly,they won't thickenuntil well into the 170-degreerange, sometimesjust before180degrees,leavinglittle time for thickeningbeforecurdling,"Corriher warnedme. "Ifthe eggsare heatedslowly, though, thickeningcanstart at150degrees and continueslowlyasthecustard heatspast160and

170 degrees."Slow, gentleheat, then, isthe

best- and probablythe only-wayto succeedwithcustards.Given thatexplanation,itwasob­viousthatcooking thecustard directly overheatwastheworstpossible way,asitheatedthecus­tardmostquickly.

Custardsciencealsoexplains why,inmynext

setoftests, Idiscoveredthatusinguptoatable­

spoonofgranulatedsugarpereggyolkimproved

6•COOK'SILLUSTRATED•MARCH/APRIL 1995

heneggsareheatedtooquicklyorwithtoohotaflame, theproteinsintheeggs,denaturedbythe heat,bondto­

gethertooextensivelyandformclumps(top),curdlingthemix­ ture.Properheatingtechniqueandthepresenceoflargesugar molecules,whichinhibitbondingtosomedegree,causethe denaturedproteins to

bondtogetherproperly (bottom)in athree-di­ mensionalmesh that

thebakedcustards tosetlike omelets. Cookingthe cus­ tardsina bainmariekeeps theirtemperaturefromrising above212degrees;this low temperatureguaranteesthat thecustardapproachesitsset point slowly andtherefore thickensgradually.Atthis lower temperature,thecus­ tards cookedin the water bathwerealsosilkier than thosebakedina350-degree oven.Asafinalrefinement,I

lowered theoven tempera­

tureto275 degreesand in­ creased thecooking timeto forty-fiveminutes.Evenbet­ ter.

Had Iexhaustedall the

custard options?Not yet!I decidedtofiddle with the temperatureof the cream. Untilnow,I'dalwaysboiled thecreamforaminuteorso and then mixed itintothe yolk-sugar mixture.Now,I triedmyrecipewithscalded

fore sprinkling them overthechilled custards. Pre-drying the brown sugarsignificantlyim­ proved itstaste, texture, and appearancewhen caramelized.Pre-drieddarkbrownsugargavethe toppingaricher flavor that wassuperior tothe light brown sugar topping,just thereverse of whenthesugarswerenotpre-dried.

It seemsthatdryingbrownsugarintheovenre­

moves moisture aswell as some ofthelumps, whichmakesiteasiertosprinkleandallowsitto coatmoreevenly. Also,sincethecaramelization processinvolvesmeltingthesugarandthenevap­ oratingsomeofitswater,havinglesswaterinthe brownsugarbeforeitisrununderthebroilerun­ doubtedly helps getthe processgoing. Adried sugartoppingneeds lesstimeunderthebroiler,so thedarkbrownsugar,withitsricherflavor,canbe usedwithoutthedanger of burning orbecoming toohard.

PERFECTCREMEBRULEE

Serves6

I tablespoonunsalted butter,softened

6 largeeggyolks,ch illed

6 tablespoons whitesugar

Jl/2 cupswhipping cream,chilled

4 tablespoonsdark brownsugar

thetextureofthecustard,whilecornstarchmade itextremelydense,grainy,andsticky.

Corriher explained that sugar moleculesare verylarge- shecalls them"Mack trucks"-and therefore come between unwound egg proteinsduringcooking, ineffectblocking,atleasttem­porarily,theirattemptstobond.Asmytestscon­firmed, addingsugarimproves thetextureofthecustard. Cornstarchworks inasimilarfashion, butunfortunatelyitalsogivesthecustardanun­pleasantgraininess.

Corriher also mentioned that stirringcon­

stantly,whichisnecessary tokeeptheheatevenly distributedinadoubleboiler, wheretheheatall comesfrom thebottom, makesthickeningmore difficult.Asyoustir,youactuallybreakapartthe eggproteinsastheyattempttobondtoeachother. Whilethisisfineforacustardlikecremeanglaise thatshould bethinenough topour,cremebrfilee hastobedense. Atthispoint,itseemed timeto moveontotheoven.

Ifirst tried placing uncooked anduncovered custardsin a warm water bath, called abain marie,inacoldoven,turned theheatto250de­ grees,andbakedforeightyminutes.Thisfirstat­ temptatoven-cookingwasadisaster.Thecustard did notsetright,cooked unevenly, andwastoo runny,and thebrownsugar toppings absorbed moisture whentheycaramelizedandturnedinto

ironplates.Morelessons learned.

1nexttriedcoveringandcookingthe custardin awarmwaterbathinapreheated,350-degreeoven forfifteenminutes.Whenthesecustardshadbeen cooked, chilled, topped andcaramelized,chilled again, andfinally served, Iknew Iwasgetting closetoreachingmygoal.

As afinaltest,I compareduncoveredcustards cookedinabainmariewiththosecookedwithout awaterbathin300-degreeoven.Dryheatcaused

cream,room temperaturecream, and chilled cream straightfromthefridge.Iwaspleasantly surprisedtofind thatthechilled cream sample wasricher,smoother,andmorevelvety thanits scaldedorroom temperaturecounterparts.

WhenImentionedthisto Corriher, shewasa

bitsurprised. Tomythinking, adding boiling or scaldedcreamtotheyolkswouldraisetheirtem­ peraturetooquickly.Corriher saidthiswascor­ rectbutthatdairyproductsareusuallyscaldedto causesomeoftheirproteins tounwindandhelp promotethickening.Aftersomethought,shesaid thiswas essential whenmakingicecream, which hasahighproportionofmilk.Unlikemilk,how­ ever,high-fatcream doesnothaveallthatmuch protein,sothebenefitsofscalding orboilingthe cream wouldbeminor.Also,addinghotcream certainlyraisesthetemperatureoftheeggsvery quickly.Sincethesecrettoperfectcremebrfilee isveryslow heat,usingchilledcreamfitinwith therest ofmyresults.

TheTopping

While working onthecustard variations,Ialso experimented withthecaramelizedsugartopping. The first recipeIhad tried called for a brown sugartoppingsothickthatitformedabarrierdif­ ficulttopenetrate withaspoon. Isoon realized that two teaspoonsof brown sugarpercreme brGhegavethebestcoverageanddepthforeven, controllable, andconsistentcaramelization.

I alsotestedtherelativemeritsoflightanddark brownsugarforthetopping.Onmyfirst try,the dark brown sugar topping burned quickly,was toohard,anddidn'ttasteasgoodasthetopping madewithlightbrownsugar.However,thelight brownsugartopping wasnotperfecteither, soI decidedtotrydryingbothlightanddark brown sugarforfifteenminutesina250-degreeovenbe-

I.Adjustovenracktocenterpositionandheat ovento275degrees.Buttersix'12-cupramekins orsix2h-cupcustardcupsandsettheminaglass

bakingpan.

2. Whiskyolksina medium bowluntilslightly thickened.Addwhitesugar andwhiskuntildis­ solved.Whiskin cream,thenpourmixtureinto preparedramekins.

3.Set bakingdishon oven rack andpourwarm waterintobakingdish tocome halfway upthe ramekins.Bakeuncovered untilcustardsarejust barelyset,about 45minutes.

4.Remove baking pan from oven,leaving

ramekinsinthehotwater;cooltoroomtempera­ ture.Covereachramekin withplasticwrap and refrigerate untilchilled, atleast2hours(canbe coveredandrefrigeratedovernight).

5.Whilecustards are cooling,spread brown sugarinasmallbakingpan;setinturned-off(but stillwarm)oven until sugar dries,about20min­ utes.Transfersugartoasmall zipper-lockfreezer bag;sealbagandcrushsugarfinewitharolling pin.Storesugarinanairtightcontaineruntilready totopcustards.

6.Adjustovenracktothenext-to-the-highest

position and heat broiler. Removechilled ramekinsfromrefrigerator,uncover,andevenly spread each with 2teaspoonsdried sugar. Set ramekins inabaking pan. Broil,watchingcon­ stantly androtatingpanforevencaramelization, untiltoppingsarebrittle,2to3minutes,depend­ ingonheatintensity.

7.RefrigeratecremebrGJeestore-chillcustard,

about30minutes.Brown sugartoppingwillstart to deteriorateinaboutI hour.•

GeneFreelandwritesaboutfoodandartcollect­

ingfromhishomeinRanchoSantaFe,California.

ILLUSTRATIONBYKARENBARNES·WOODRONSAVILLEHARLIN,INC.

MARCH/APRIL. 1995•COOK'SILLUSTRATED •7