Composition and Properties of Aquafaba: Water Recovered from Commercially Canned Chickpeas

My name is Martin Reaney. I'm Chair of Lipid Quality and Utilization at the University of Saskatchewan, Department of Agriculture and Bioresources. Pulses and pulse fractions have been recently used as milk and egg replacements.

For this purpose, we investigated the use of aquafaba for milk and egg replacement. Aquafaba is the water remaining from canned chickpeas. This water contains components that can thicken food products and stabilize food foams and emulsions.

There are many recipes on the internet which describe the use of aquafaba in food products. However, people often report that they could not repeat recipes that they found on the internet and achieve the same properties of aquafaba that other users have achieved. The overall goal of our study was to investigate the correlation between aquafaba composition, density, viscosity and foaming properties.

NMR and electrophoresis were used to study aquafaba composition. There is a procedure used by all manufacturers producing canned chickpeas. This involves first sorting the seed for quality to remove undesirable seed.

Optionally, the seed is rinsed with water to remove surface contamination. For our purposes, the seed is submerged in water for a few seconds and then the water is immediately drained. After the seed is rinsed, the seed is immersed in two volumes of water or more to allow it to expand and fully hydrate.

This soaking is called steeping. After soaking, the seeds have expanded and are ready to can. The seeds are comparatively softer and more fragile and must be handled appropriately.

This involves, first of all, sorting the seeds for quality to remove undesirable seeds, then it follows that the seeds would be soaked in water. First they're rinsed and then soaked for a period of time, as you can see here. After soaking, they will have greatly expanded in volume and they are ready for canning.

After steeping the seeds, we pour the soaked seed through a sieve. Water going through the sieve is discarded. The seed is transferred to a 300mL can with a large spoon such as this one here.

The can is filled with water to within one centimeter of the top so that sufficient room is available for expansion. The canning lid was placed on top of the can and placed in the can sealer as demonstrated. The sealer is operated by pressing down on the handle to force the lid into contact with the can during sealing.

The hand crank on the sealer is turned to both spin the can and to seal the can with two die that shape the lid and fix it to the can. Once the can is sealed, the die retracts and the can is readily removed. The can's seal should be watertight.

Then, the sealed can is placed in a pressure cooker. Water is added to the cooker to fill it to about four centimeters deep. The cooker is sealed by placing the lid on the cooker and rotating it until it is sealed as shown.

For pulses, the cooking time is a period of 75 minutes at 115 degrees centigrade. After cooking the peas, we have to take them out and separate the peas from the aquafaba. We open this with a standard can opener.

Now that the can is open, we can place the contents of the can in a colander that is put in a larger collection vessel. The aquafaba juice goes through the colander and can be separated that way from the peas in the can. It's this liquid that we use for further experiments.

So, that's aquafaba of two minutes. Here, you can see the type of foam we would use for further studies. In this study, we used commercial cans from 10 different commercial sources as our sources of aquafaba.

After blending for two minutes, the foam was placed in plastic cups and observed over time. As you can see, the foam is collapsing in all the cups. Only two samples, D and H, produced a stable foam that lasted throughout the entire experiment.

All of the other foams collapsed over time. At zero hours, all cups were full of foam. After one hour, separation of liquid was evident in all samples except Sample H.After 14 hours, some of the liquid had separated from Sample H.We further wanted to characterize aquafaba from canned peas.

We looked at gross characteristics, mass balance, including looking at the content of chickpea seeds, juice volume, juice density and juice viscosity. The correlations among these properties were also investigated. Unsurprisingly, there was a strong negative correlation between chickpea fresh weight and juice volume.

There was a strong positive correlation between chickpea fresh weight and aquafaba viscosity. Aquafaba samples were prepared for NMR spectroscopy. The aquafaba, shown here as juice, was centrifuged of 9200g in an Eppendorf tube for 10 minutes.

The supernatant was taken up from the Eppendorf tube in a one mL slope syringe as seen here. Then about four or five micron filters attached to the Luer-Lok on the syringe. We found that the filtration required some application of force to expel the sample.

Once the sample is filtered, it is possible to use it for NMR spectrometry. Here, the sample is being transferred to an NMR tube. Later, deuterated water will also be added to the sample to allow the NMR spectrometer to lock to its frequency.

Based on the NMR resonances observed, it is possible to determine the compounds that are present in aquafaba. We were able to observe compounds that reflected conditions experienced by the chickpea during processing. For example, conditions during steeping might allow anoxic respiration by the seed or microorganisms present.

We observed compounds such as lactic and acetic acid that are likely formed during steeping. We also observed compounds that likely leach from the seed during cooking. These includes compounds such as choline and sugars.

The supernatant was also used for peptide mass fingerprinting after removing low molecular weight species. In this case, the chickpea liquid was placed in a centrifugal unit with a three kilodalton membrane. The centrifugal unit was used to centrifuge it at 3900 times g for two hours.

The retentate of this centrifugation was used for SDS-PAGE analysis. After standard separation of the proteins by SDS-PAGE, the gel was stained using Coomassie Blue. Intensely stained bands were then cut from the gel, destained, subject to tryptic digestion and the digestion fragments were analyzed using LC-mass spectrometry, Protein mass spectrometry software was used to deduce the sequences of proteins present.

Most of the proteins present appear to be chickpea, but a few high molecular weight proteins are present that are due to microbial contamination from the chickpea disease organism, Ascochyta blight. The majority of proteins identified are well-known, heat stable proteins, including dehydrants, late embryo abundant proteins and defensins. Additives to chickpea include salt, disodium ADT and calcium chloride.

The can linings that we observed were white, clear yellow or metallic. Chickpea fresh weight is positively correlated to its juice viscosity and the VF/VJ ratio. This relationship is largely related to the samples without an AZL added.

These samples have a much higher viscosity, but much lower VF/VJ than other samples. Chickpea fresh weight and juice volume are easily determined by consumers. We observed a negative correlation between juice density and volume increase ratio.

The foam formed from aquafaba samples derived from chickpeas cooked with salt had much greater stability. Aquafaba contains proteins with molecular weights that are greater than three kilodaltons. End of our analysis shows the presence of substances in chickpeas, for example choline, and also compounds that might accumulate during soaking the peas due to action of bacteria or seed metabolism, such as acetic acid and lactic acid.

Electrophoresis and peptide mass fingerprinting has allowed a preliminary characterization of proteins present.