Tuesday, March 13, 2012

Replacing Butter: Vegetable Oil

It's been a while since I've been able to post something here on the blog, but I promise I have a good reason for it. Last month, I moved from Washington, DC (my home of the last few years) to New York City as part of a new job. I was so busy trying to get everything together, I just didn't have the time to bake or blog. 

But now I've been here for a couple of weeks, and last week I kicked off the new chapter of crazy baking. I began with some hamentashen for Purim, then made a gingerbread cake for Sabbath. I was all set to spend some time on Sunday making some cookies when an idea hit me. Now would be a good a time as any to revisit a problem that had been plaguing me for many years: how to replace melted butter in a cookie recipe.

[I suppose now is a good a time as any to alert you that there's going to be a lot of science talk in this post. Hold on to your hats. Also, it's worth noting that I relied heavily on Harold McGee's "On Food And Cooking" for the science aspect of this. It's an invaluable resource.]

As I've said before in other posts, when I make cookies for Sabbath or holidays, I usually need to keep them pareve, which means they have to be dairy-free (and meat free, but who's putting meat in a cookie?). The usual solution is to grab a stick of dairy-free margarine and substitute it for the butter. I've never been a big fan of this for a couple of reasons:

1) Margarine is pretty gross. 
2) Pareve margarine isn't so easy to find (at least back in DC).
3) Margarine is not so healthy for you. 

Let me take a moment to address that last one. While doing some research ahead of this post, I discovered that, for a while, margarine was thought to be healthier than butter. Butter is high in saturated fat (not heart healthy) while margarine is high in unsaturated fat (more heart healthy, relatively speaking). But more recently, scientists discovered that the process by which oil is made solid for margarine, known as hydrogenation, turns those healthier unsaturated fats into trans fats. Trans fats are just as bad as saturated fats, and in certain ways, somewhat worse. (For more information about this, I'd recommend the American Heart Association page about fats and oils, with thanks to Leah McGrath, R.D. for pointing me towards the source. You should follow her on Twitter at @InglesDietitian)

For all those reasons, I decided to see how well I could replace butter using liquid vegetable oil as the fat, thus preserving the unsaturated fats and being a heck of a lot more convenient to boot. I don't know about you, but I've always got a bottle of vegetable or canola oil in my pantry, while my margarine is very single purpose (and did I mention gross? It smells like apples and I can't tell why!). 

Taking a page out of my creaming method experiment, I decided I would take a cookie recipe that used melted butter, make a batch, then make a second batch with my substitution of vegetable oil. Then, I would get people to try one of each cookie and tell me what they thought. I would use that information to draw some kind of conclusion, although it would never stand up to rigorous scientific scrutiny.

I chose America's Test Kitchen's Molasses Spice Cookies for this test; somewhat because I wanted my first NY post to pay tribute to my first DC post, but mostly because I like the recipe. I made the first batch without any problems then moved on to the second batch. I checked the recipe and saw that it needed 12 tablespoons of melted butter.  "Great," I can hear you saying. "Just put in 12 tablespoons of oil and be done! Right?"


Interesting fact: Butter is required by law to be about 80% fat by weight. Look at any box of butter and you should see that for every 14 gram serving, only 11 grams are fat. Approximately 16-18 percent of the remaining weight is water (depending on who you ask) and the rest are milk solids and basic elements. The inclusion of water makes a difference in the final end product as well, as it leads to steam which gives some extra puffiness to the cookie during baking. Therefore, 12 tablespoons of oil was not the right way to go here.

Back when I made the first batch of cookies, I weighed the butter on my scale and got 172 grams. Using the ratio of 11 grams of fat per 14 grams of weight, I came up with about 135 grams of fat. Accounting for some rounding errors, I settled with 136 grams of oil. The water was a bit trickier. I was aiming for 32 grams of water, which would have been 18.6% by weight. Unfortunately, I over-poured a bit and ended up with 34 grams of water (19.7% by weight). I made a note to add a bit of flour later on to account for the extra liquid. 

I proceeded with making the cookies without any other incidents. One thing I did note was the difference in the consistency of the cookie batter. The butter batter was much stiffer and drier, whereas the oil batter was very loose and sticky. This wasn't particularly unexpected though. Butter is a semi-solid at room temperature, with some of the solid fat turning into liquid (and back again if the temperature is right) and other fat particles remaining in the liquid state once it's reached. So with time and the right temperature conditions, the butter-based batter would tend to be a bit firmer.

When I removed the oil-based cookies, I noticed that they were a lot flatter than I expected. I couldn't figure out what had happened. I thought i had accounted for the puff by adding the water! It wasn't until a couple of hours later that I realized what the culprit might have been: the lack of acid. Milk, believe it or not, is an acid; primarily lactic acid. That acid reacts with baking soda to produce lift and puff in the cookie. Oil, on the other hand, has no pH value and is therefore neutral. I wasn't getting as much of a reaction as I wanted from my leavening agents! I'm not sure what the best way to fix that would be, but that's for another experiment.

"Enough with the science!" I can hear you shouting. "What about the TASTE?"

OK, OK! I gave four people one of each cookie and asked them to tell me which they preferred and any other comments (A fifth person was given cookies as well, but I lost track of which cookie was which so I had to invalidate her data point. Sorry, Rebecca.). Three out of four preferred the oil cookie, praising its chewiness. Everyone agreed that the texture of the oil cookie was softer than the butter cookie (one called it "mushy"). Two people said the butter cookie was sweeter, which I could understand given the presence of lactose (milk sugar) in butter. Overall, everyone thought both cookies were good and they'd eat both again.

So what's the takeaway from this experiment? I think my oil and water solution gets me about 80% of the way towards a good replication of the butter cookie. I still need to find a way to introduce a bit more acid to work with the leavening agents. I also would like to see what would happen if I used a bit of shortening to provide more structure to the fat, since shortening is a solid at room temperature. It would negate a lot of the "health benefits," but I could live with that.  Broadly, I think the takeaway is that using the oil to replace the fat proportionally by weight works very well and is worth some more exploration in the future. 

Oh, and I guess the other takeaway is: Either way, you've still got a darn good cookie to eat.