Media Releases

Want ripples on your icicles? University of Toronto scientists suggest adding salt

October 10, 2013

TORONTO, ON – Though it’s bare­ly the begin­ning of autumn, sci­en­tists at the Uni­ver­si­ty of Toron­to are one step clos­er to explain­ing why winter’s ici­cles form with Miche­lin Man-like rip­ples on their elon­gat­ed shapes.

Exper­i­men­tal physi­cist Stephen Mor­ris and PhD can­di­date Antony Szu-Han Chen were spurred to inves­ti­gate by the rip­ples that appear around the cir­cum­fer­ence of ici­cles that occur nat­u­ral­ly.  It has been the­o­rized that the rip­ples are the result of sur­face ten­sion effects in the thin water film that flows over the ice as it forms. Their inves­ti­ga­tion revealed that the actu­al cul­prit is salt.

“Nobody has sys­tem­at­i­cal­ly inves­ti­gat­ed what caus­es the rip­ples so we began grow­ing them in the lab,” said Chen, lead author of a paper pub­lished online this week in New Jour­nal of Physics. Account­ing for key fac­tors that influ­ence the shape of an ici­cle as it forms in nature – ambi­ent tem­per­a­ture, flow rate of water and the motion of the air sur­round­ing it – the researchers exper­i­ment­ed with the com­po­si­tion of the source water.

“We had already tried Toron­to tap water and found that it formed rip­ply lab­o­ra­to­ry ici­cles, when dis­tilled water didn’t,” said Mor­ris. “We also con­firmed that melt­ed rip­pled ici­cles tak­en from Toron­to garages were very slight­ly salty, so that’s what led us to pur­sue the com­po­si­tion fac­tor.”

Using pure dis­tilled water, dis­tilled water with small quan­ti­ties of sodi­um chlo­ride added, and Toron­to tap water – which con­tains sodi­um chlo­ride as well as many oth­er impu­ri­ties – they pro­duced 67 sam­ples grown under a broad range of con­di­tions. The evo­lu­tion of the ici­cle shapes over time was acquired from dig­i­tal images using detec­tion of their edges, which were then ana­lyzed with com­put­er image pro­cess­ing.

Rip­ple growth was not observed on dis­tilled water ici­cles, where­as salti­er ici­cles showed clear rip­ples that appear in a patchy way and some­times grew as large as a few mil­lime­tres. The rip­ples were seen to move slow­ly upward dur­ing the ici­cle growth, though the researchers note that both the speed and direc­tion of the rip­ple motion could vary depend­ing on the con­cen­tra­tion of dis­solved salt.

Mor­ris and Chen found that rip­ples only became appar­ent at the remark­ably low salin­i­ty of the water with 20 mg of salt per litre. This lev­el, in fact, is a con­sid­er­ably low­er lev­el of impu­ri­ty than found in com­mon tap water.

“We even added a non-ion­ic ingre­di­ent to the dis­tilled water to reduce the sur­face ten­sion of the thin film of water flow­ing over the ici­cle, and it didn’t pro­duce rip­ples,” said Chen. “Instead, rip­ples emerge only on ici­cles grown from water with dis­solved ion­ic impu­ri­ties.”

“Our moti­va­tion is pure curios­i­ty about nat­ur­al pat­terns, but the study of ice growth has seri­ous appli­ca­tions, includ­ing ice accu­mu­la­tion on air­planes, ships and pow­er lines,” said Mor­ris. “This result is total­ly unex­pect­ed, not just by us before we did this, but by the­o­rists and exper­i­men­tal­ists in our field who study ice dynam­ics and pat­tern for­ma­tion.”

“No the­o­ry accounts for the effect of salt, so the shape of ici­cles and the rea­son for their rip­ples are still mys­ter­ies. Except we now know that a lit­tle salt is required in the recipe.”

The find­ings are report­ed in the paper “On the ori­gin and evo­lu­tion of ici­cle rip­ples”. The research is sup­port­ed by fund­ing from the Nat­ur­al Sci­ences and Engi­neer­ing Research Coun­cil of Cana­da.

- 30 -

Note to media: Vis­it for images and videos illus­trat­ing the research described here.

For more infor­ma­tion, con­tact:

Stephen Mor­ris
Depart­ment of Physics
Uni­ver­si­ty of Toron­to

Antony Szu-Han Chen
Depart­ment of Physics
Uni­ver­si­ty of Toron­to

Sean Bet­tam
Com­mu­ni­ca­tions, Fac­ul­ty of Arts & Sci­ence
Uni­ver­si­ty of Toron­to