[Bill_Toge]

@Margreat, in my experience SLS is used in toothpastes between 0.5 and 1.5%, and it’s best to dissolve in part of the water before you add it to the batch; you may need to heat the water to achieve a full and timely dissolution

you’ll need to mix it into the batch slowly and in stages, as it foams a lot and makes a the product more prone to aeration

@David, you only use CPC at 0.1-0.2% MAXIMUM, so the inherent hazards of the raw material are considerably reduced in the final product

[Bill_Toge]

@Belassi, @manstra, yes, aqueous solutions of sorbic acid (the active form of potassium sorbate) are easily oxidised by oxygen, forming a multitude of by-products

further details can be found in these papers - the latter identifies beta-carboxyacrolein as the species responsible for browning:

http://pubs.acs.org/doi/abs/10.1021/jf60232a072?journalCode=jafcau

http://www.sciencedirect.com/science/article/pii/030881468890074X

[Bill_Toge]

with phenoxyethanol and benzyl alcohol you risk them separating, since both have low solubility, even in surfactant systems

potassium sorbate is most effective below pH 6, but causes terrible yellowing/browning in liquid products

in my view, DMDMH and sodium benzoate are the only ones which are efficacious and practical for a shampoo; and you’ll have to keep your pH below 5 for the sodium benzoate to be effective

[Bill_Toge]

interestingly, we had a query from the Daily Mail several years ago about making permanent hair colours for them

it collapsed over an argument about the text they wanted to use as pack copy; like most of their publications, it was misleading at best and just plain wrong at worst, and we were extremely unwilling to accept it without some drastic editing

[Bill_Toge]

@Perry, the Daily Mail is notorious for its hysterical sensationalism, its tendency to manufacture moral outrage, and its lack of concern for factual accuracy; most of the time it just wants to scare readers and get them angry about an issue rather than properly informing them about it

[Bill_Toge]

this is the legally prescribed wording for conditions of use and warnings for oxidative colourants containing PPD (EU Cosmetic Product Regulations Annex III/8, emphasis mine):

To be printed on the label: The mixing ratio. “ Hair colourants can cause severe allergic reactions. Read and follow instructions. This product is not intended for use on persons under the age of 16. Temporary ‘black henna’ tattoos may increase your risk of allergy. 

Do not colour your hair if: 

— you have a rash on your face or sensitive, irritated and damaged scalp, 

— you have ever experienced any reaction after colouring your hair, 

— you have experienced a reaction to a temporary ‘black henna’ tattoo in the past. 

Contains phenylenediamines. Do not use to dye eyelashes or eyebrows” 

many branders, L’Oreal among them, also instruct the user to do a patch test 48 hours before using hair dyes, and not to use it if they have an allergic reaction

in my view, if any action is required above and beyond the existing measures, I’d suggest putting “READ INSTRUCTIONS BEFORE USE” in big red letters all over the pack, so even the densest and least observant consumers can’t miss it

(also, it’s not true that PPD is “required” for dark colours: toluene-2,5-diamine sulphate is a less allergenic alternative to PPD which has been available for years)

[Bill_Toge]

I’m a little confused here, are you using MEA or Comperlan 100 (cocamide MEA)? 

if the latter, it’s very likely to be the culprit; it’s only just soluble in surfactants at high temperatures, and often crashes out at room temperature or below unless you use a polymer capable of suspending solids 

(at room temperature, MEA on its own is an extremely pungent liquid which is more likely to thin a shampoo than thicken it)

[Bill_Toge]

Euxyl PE9010 is an exceptionally reliable preservative, but it’s better suited to emulsions and you’ll have a lot of trouble dissolving it in a gel

when DMDMH has failed on its own, we’ve used Glydant Plus from Lonza, which is a blend of DMDMH and iodopropynyl butylcarbamate; it’s never failed us yet

[Bill_Toge]

in my view you’ve got a good formula there, but you’ve just got to find the best way of making it

I can’t stress enough that processing is a HUGE contributing factor for W/O emulsion stability - for best results, form the emulsion at as low a temperature as you can get away with, use a good shear mixer with a fine screen on the stator head, and be sure to make your batches consistently every time

personally, if I were to add anything it’d be hydrogenated castor oil (Cutina HR from BASF, also available from Vertellus and a few others) - it’s a rheology modifier which helps to stabilise it at higher temperatures

in my experience, 1% is usually sufficient

[Bill_Toge]

agree with @Kirk - homosalate (a UV filter) is the salicylate ester of menthol, and that smells nothing like mint either

and if it were methyl acetate, it’d have a very strong smell like glue or wood varnish

[Bill_Toge]

from a microbiological point of view, the big problem with xanthan gum is that it’s an untreated polysaccharide, so it sustains microbial growth

cellulose thickeners are synthetically altered, and can’t be digested by microbes, so the risk factor is much lower

[Bill_Toge]

@Lara, glycerine is a 1,2,3-triol rather than a glycol, and as @Belassi said, it is not antimicrobial in itself; however, it can be used to reduce the available water and hinder microbial growth

(the catch: you have to use it at 40-50% w/w or higher for it to have any significant effect)

on the other hand, 1,2-diols (propylene, pentylene, caprylyl glycol etc.) DO show some antimicrobial activity, which increases with the length of the carbon chain

[Bill_Toge]

without seeing the product in question, I’d guess the “stringy texture” comment is related to the rheological properties of the resin used for setting the hair, which can have a significant effect on the properties of the product as a whole

for instance, PVP can make a gel feel sticky to the touch if it has a high K-number (90 or 120)

other examples of sticky-feeling resins are Aquaflex XL-30 and Advantage LC-E; the former is compatible with carbomers, so it could in principle be used in a gel

have you got pack copy and an INCI list for this product?

[Bill_Toge]

as I see it, the root cause of your problem is that phenoxyethanol is soluble in water at 80°C, and hardly soluble at all at room temperature

so when your emulsion is formed, it initially partitions into to the water phase, but as it cools down some of it migrates into the oil

when I’ve used Euxyl PE9010 in W/O products, I’ve always added it after the emulsion has been formed, for that very reason, and I’ve not had a problem with it yet

(incidentally, 80°C seems excessively high for this type of product; a good way to reduce bleeding would be to combine phase A and B at a temperature that’s above the drop point of your oil phase, but otherwise as low as possible, since the same phenomenon affecting the phenoxyethanol can affect your emulsifiers too)

hope this helps!

[Bill_Toge]

first things first: what’s in your oil phase, how much water is there, which emulsifiers are you using, and what temperature are you forming your emulsion at?

also, when you say it’s breaking, is it bleeding, or is it separating into two layers?

[Bill_Toge]

since you’ve got no strong acids or strong bases in the formula, the chance of hydrolysing the nicotinamide (or any amide, for that matter) is low

this page describes the general process of amide hydrolysis

[Bill_Toge]

likewise, the company I work for is based in Europe and our site is not FDA accredited, so I can’t help you there

there are very few manufacturers who make hair colours, as they are messy/difficult products to  handle in bulk; good luck in your search!

[Bill_Toge]

try http://www.efchemicalconsulting.co.uk

in my experience their turnaround times are prompt, and they’re pretty sensible; some of the bigger companies can be incredibly anal and pedantic

[Bill_Toge]

sorry, I misread the thread originally and somehow thought there was water in this formula

I agree with @Iaskedbetter, the pH reading in an anhydrous product is totally irrelevant; reference electrodes are only meaningful or useful when measuring the pH of water/solvent mixtures

[Bill_Toge]

a paddle stirrer like the ones you’ve linked to will not do the job of a high-shear mixer, no matter how fast it goes; and to create reasonably stable creams, you will need a high-shear mixer

the GM-A or GM-B model lab mixer from Joshua Greaves & Sons is a good general-purpose mixer for the scale you want; last time we bought a GM-B it was about £1,300

http://www.greaves.co.uk/pdf/GM%20mixers.pdf (lab mixers are on page 9)

though the brochure states the capacity of the GM-B is 2-15 litres, the minimum batch size in practise is about 300 ml (the mixer head won’t physically fit into anything smaller)

[Bill_Toge]

where in the world are you based, and where are you intending to sell your products?

[Bill_Toge]

solvent-based nail polishes typically contain quaternised bentonite/hectorite clays as suspending agents, e.g. Bentone 27V CG from Elementis (INCI: stearalkonium hectorite)

[Bill_Toge]

the melting point is for the pure substance, not a solution

if it’s reasonably soluble in the oil you want to dissolve it in, it’ll dissolve at a much lower temperature than the melting point

I agree with @Belassi; since it’s an ester, try dissolving it in another ester

[Bill_Toge]

if the flavour is IFRA compliant and has a certificate confirming this, then “Aroma” (in Europe) or “Flavor” (in the US) would be sufficient

also, for labelling purposes, it would be helpful to know how much of it is sunflower oil, and how much of it is flavour

[Bill_Toge]

yes, I figured you were making that type of product

as I said, try adding microcrystalline wax with a high melting point, and/or lanolin wax; either of them will give you a good strong hold and allow you to keep your formula and process simple

if you try to turn it into a water-in-oil emulsion, you’ll only make a rod for your own back