lable at ScienceDirect

Food Microbiology 27 (2010) 33–36

Contents lists avai

Food Microbiology

journal homepage: www.elsevier .com/locate/ fm

Evaluation of the efficacy of four weak acids as antifungal preservativesin low-acid intermediate moisture model food systems

Yang Huang*, Mark Wilson, Belinda Chapman, Ailsa D. HockingCSIRO Food and Nutritional Sciences, 11 Julius Avenue, Riverside Corporate Park, North Ryde, NSW 2113 Australia

a r t i c l e i n f o

Article history:Received 2 February 2009Received in revised form17 July 2009Accepted 20 July 2009Available online 24 July 2009

Keywords:Weak acidFood spoilage fungiMinimal inhibitory concentrationUndissociated form

* Corresponding author. CSIRO Food and NutritionaRyde, NSW 1670, Australia. Tel.: þ61 2 9490 8527; fa

E-mail address: yang.huang@csiro.au (Y. Huang).

0740-0020/$ – see front matter Crown Copyright � 2doi:10.1016/j.fm.2009.07.017

a b s t r a c t

The potential efficacy of four weak acids as preservatives in low-acid intermediate moisture foods wasassessed using a glycerol based agar medium. The minimum inhibitory concentrations (MIC, % wt./wt.) ofeach acid was determined at two pH values (pH 5.0, pH 6.0) and two aw values (0.85, 0.90) for five foodspoilage fungi, Eurotium herbariorum, Eurotium rubrum, Aspergillus niger, Aspergillus flavus and Penicilliumroqueforti. Sorbic acid, a preservative commonly used to control fungal growth in low-acid intermediatemoisture foods, was included as a reference. The MIC values of the four acids were lower at pH 5.0 thanpH 6.0 at equivalent aw values, and lower at 0.85 aw than 0.90 aw at equivalent pH values. By comparisonwith the MIC values of sorbic acid, those of caprylic acid and dehydroacetic acid were generally lower,whereas those for caproic acid were generally higher. No general observation could be made in the caseof capric acid. The antifungal activities of all five weak acids appeared related not only to the undisso-ciated form, but also the dissociated form, of each acid.

Crown Copyright � 2009 Published by Elsevier Ltd. All rights reserved.

1. Introduction

Weak acids are commonly incorporated into food and feed toincrease shelf life by preventing the growth of microorganisms. Themost common weak acid preservatives used in foodstuffs are sorbicacid, benzoic acid and propionic acid. Of these sorbic acid, or morecommonly sodium or potassium sorbate, has been reported to havethe greatest antifungal activity at pH values up to 5.6 (Ray andLiewen, 2004) and to be one of the most effective compounds forpreventing fungal growth in intermediate moisture foods (Guynotet al., 2005; Suhr and Nielsen, 2004). A number of acids and acidderivates not commonly used in low-acid intermediate moisturefoods, including short chain saturated fatty acids (C6 - C18) anddehydroacetic acid, have also been reported to exhibit antimicro-bial activity in laboratory media (Kabara et al., 1972; Kabara andMarshall, 2005; Kato and Shibasaki, 1975; Morozumi et al., 1985;Rihakova et al., 2001; Skrivanova et al., 2005). Malfeito Ferreiraet al. (1997) indicated that caproic acid (C6) and caprylic acid (C8)were more active against yeasts than acetic, benzoic, butyric andpropionic acids when compared on a molar basis (w1–2 mM at pH4.0), while Kato and Shibasaki (1975) reported that the minimuminhibitory concentrations (MIC) of capric acid (C6) and caprylic acid

l Sciences, PO Box 52, Northx: þ61 2 9490 8581.

009 Published by Elsevier Ltd. All

(C8) were 1 and 2–4 mM, respectively, for fungi grown on CzapekDox agar (pH 5.6). Morozumi et al. (1985) reported that the inclu-sion of 0.05% dehydroacetic acid in culture media inhibited thegrowth of a number of moulds and that a greater inhibitory activityon a wt./wt. basis was observed with dehydroacetic acid than withsorbic acid, potassium sorbate or sodium benzoate. Kato and Shi-basaki (1975) reported similar findings in a study of the inhibitionof growth of Aspergillus niger, indicating that the MIC of dehydro-acetic acid was 10-fold lower than that of sorbic acid. Unfortunately,the publications discussed lack information relating to the aw of thesystems employed. Moreover, the growth periods over which theantimicrobial activity of the acids was determined were relativelyshort. To ascertain the potential of these weak acids to inhibit thegrowth of fungi in intermediate moisture foods, those issues needto be addressed.

In low-acid environments (pH > 4.6) the dissociation of weakacids is incomplete with a substantial amount existing in a disso-ciated form. Although it is generally accepted that weak acidsexhibit a higher degree of antimicrobial activity in the undissoci-ated form (Stopforth et al., 2005), it has been demonstrated that thedissociated forms can contribute to the antimicrobial activity, andparticularly to the antifungal activity of sorbic acid (Eklund, 1983;Skirdal and Eklund, 1993).

In the present study, we determined the MIC of caproic acid,caprylic acid, capric acid and dehydroacetic acid, as well as that ofsorbic acid, under conditions of intermediate moisture (0.85 aw and0.90 aw) and low-acid (pH 5.0 and pH 6.0) over an extended

rights reserved.

Table 1Concentrations (%, wt./wt.) of weak acids used for determination of the minimuminhibition concentrations (MIC).

acid/sodium orpotassium salt

pH aw acid equivalent concentration (%, wt./wt.)

caproic acid 6.0 0.85 0.4, 0.3, 0.2, 0.1, 0.050.90 0.5, 0.45, 0.425, 0.4, 0.3, 0.2, 0.1, 0.05

5.0 0.85 0.3, 0.2, 0.1, 0.05, 0.010.90 0.45, 0.4, 0.3, 0.2, 0.1

sodium salt ofcaprylic acid(sodium octanoate)

6.0 0.85 0.25, 0.2, 0.15, 0.1, 0.050.90 0.25, 0.2, 0.15, 0.1, 0.05

5.0 0.85 0.2, 0.15, 0.1, 0.05, 0.010.90 0.2, 0.15, 0.1, 0.05, 0.01

sodium salt ofdehydroacetic acid

6.0 0.85 0.125, 0.1, 0.075, 0.05, 0.010.90 0.2, 0.15, 0.125, 0.1, 0.075, 0.05, 0.01, 0.001

5.0 0.85 0.125, 0.1, 0.075, 0.05, 0.010.90 0.125, 0.1, 0.075, 0.05, 0.01

potassium sorbate 6.0 0.85 0.4, 0.35, 0.3, 0.25, 0.2, 0.15, 0.10.90 0.55, 0.5,0.45, 0.4, 0.3, 0.2, 0.1

5.0 0.85 0.3, 0.25, 0.2, 0.15, 0.1, 0.05, 0.010.90 0.3, 0.25, 0.2, 0.15, 0.1, 0.05, 0.01

sodium salt of capric acid 6.0 0.85 0.2, 0.15, 0.1, 0.075, 0.050.90 0.2, 0.15, 0.1, 0.075, 0.05

5.0 0.85 0.15, 0.1, 0.075, 0.05, 0.010.90 0.15, 0.1, 0.075, 0.05, 0.01

Y. Huang et al. / Food Microbiology 27 (2010) 33–3634

(12 week) incubation period using agar-based systems. The MICvalues of the four weak acids were compared with those of sorbicacid on both a weight for weight (wt./wt.) basis and a molar basis,and the antifungal activities of both the dissociated and theundissociated forms were analysed.

2. Material and methods

2.1. Preparation of media

The model system for low-acid intermediate moisture foods wasbased on dichloran-glycerol agar (Oxoid CM729, Oxoid AustraliaPty Ltd, Adelaide, South Australia) and comprised 0.5% peptone, 1%glucose, 0.1% KH2PO4, 0.05% MgSO4, 0.01% chloramphenicol, and1.5% agar on a weight for total weight basis (% total wt.). The aw ofthe base medium was adjusted to 0.90 and 0.85 by maintaininga constant ratio of glycerol (Selby Biolab, Australia) to water: 0.450for 0.90 aw and 0.785 for 0.85 aw. Due to poor solubility of the acidform, the sodium salts of caprylic acid (sodium octanoate; Sigma C-5038, USA), capric acid (Sigma C-4151, USA), and dehydroacetic acid(Merck 8.14085.0250, Germany) and the potassium salt of sorbicacid (Sigma 85 520, Germany) were used. Caproic acid (Aldrich153745, USA) was sufficiently soluble in its acid form. Acids or theirsalts were added to the model system to yield final concentrationsof 0.01–0.55% (wt./wt.) (Table 1). The components of each mediumwere mixed with deionised water and the media were then steri-lised by steaming for 45 min. After steaming, the pH was adjustedwith 5 M KOH or 10 M HCl to pH 5.0 or pH 6.0. Aliquots (20 ml) ofthe media were subsequently poured into Petri dishes, and allowedto solidify and dry at room temperature overnight. After solidifi-cation, the aw was checked with an Aqualab CX-3 instrument(Decagon, Pullman, WA, USA), and the surface pH was assessedwith a Beckman pH meter equipped with a surface pH probe(510066, Beckman Instruments, USA).

2.2. Preparation of inocula

Five fungal isolates from the FRR culture collection of FoodScience Australia were used in the present study. Eurotium

herbariorum (FRR 5354) and Eurotium rubrum (FRR 3613) weregrown on Czapek yeast extract 20% sucrose agar (CY20S) at 25 �Cfor up to three weeks. Aspergillus flavus (FRR 5660), A. niger (FRR5664) and Penicillium roqueforti (FRR 3063) were grown at 25 �C onCzapek yeast extract agar (CYA) for one week. CY20S and CYA wereprepared in the manner of Pitt and Hocking (1997). Once sporula-tion had occurred spores suspensions of each species wereprepared as follows. For Eurotium species, colonies were wet withsterile 0.05% Tween 80 solution, and scraped off the plates usinga small spatula to make a paste. The paste was transferred ontoa microbiological slide, and rubbed with another slide 10–15 timesto release the ascospores from the asci. The spores were thencollected in sterile 28 mL tubes with 2–3 g of glass beads (3 mmdiameter) and resuspended in sterile deionised water. For Asper-gillus and Penicillium species, the colonies were flooded with 2 ml ofsterile 0.05% Tween 80 solution, scraped off with a spatula, andcollected into sterile 28 ml tubes with glass beads (3 mm indiameter). The spore/bead suspensions were vortexed for 1 minand filtered through sterile glass wool for the removal of hyphaeand beads. The filtered spore suspensions were then diluted insterile 50% glycerol to a final concentration of 6 � 104 cfu/ml,dispensed into sterile 1 ml Cryogenic vials at aliquots of 200 ml, andstored at �80 �C.

2.3. Determination of minimum inhibitory concentration (MIC)

Duplicate Petri dishes containing each of the media were three-point inoculated with 5 ml of each spore suspension (ca. 300 sporesper spot). After inoculation the plates were left on the bench for30 min to dry, then placed in polyethylene boxes with air-tight lidsand incubated at 25 �C for up to 12 weeks, with incubation ofa single species in the presence of single acid per box. The aw of themedia was controlled by placing saturated solutions of Sr(NO3)2

(aw ¼ 0.850) or BaCl2 (aw ¼ 0.902) in the boxes. The inoculatedmedia were examined daily or weekly with a stereomicroscope forvisible fungal growth. The experimental MIC (% total wt.) wasdetermined as the lowest concentration of each acid at which novisible growth of the fungi was observed for all six points after12 weeks incubation.

2.4. Calculation of MIC of total acid and MIC of undissociated form

Both undissociated and dissociated forms of a weak acid coexistin water solutions. Eklund (1983) introduced the symbol ‘a’ as theratio between undissociated and total acid in a solution, andcalculated the value of ‘a’ using the pKa of the acid and the pH of thesolution by equation (1):

a ¼ 1=�

10pH�pKa þ 1�

(1)

MIC values on the molar basis of total acid in the aqueous phase(MICt, mmol/l) were estimated from the experimental MIC (% totalwt.), taking into account the added water used in the agar media,i.e. 66.6 (% total wt.) for media with 0.90 aw and 54.0 (% total wt.) formedia with 0.85 aw. Based on equation (1), the MIC values for theundissociated form in the aqueous phase (MICu, mmol/l) werecalculated as shown in equation (2), using pKa values of 4.85 forcaproic acid (Lide and Frederikse, 1997), 4.89 for caprylic acid (Lideand Frederikse, 1997), 4.90 for capric acid (Mukaiyama et al., 1954),5.27 for dehydroacetic acid (Doores, 2005) and 4.76 for sorbic acid(Eklund, 1983).

MICu ¼ MICt=�

10pH�pKa þ 1�

(2)

Table 2Minimum inhibitory concentration (MIC, % total weight) of weak acids for five fungalspecies.

Condition Species MIC (%, wt./wt.) of weak acid

sorbic dehydroacetic caproic capric caprylic

pH 6.0 0.85 aw E. herbariorum 0.15 0.1 0.2 0.075 0.1E.rubrum 0.2 0.075 0.3 0.075 0.15A. flavus 0.1 0.05 0.05 0.05 0.05A. niger 0.3 0.05 0.1 0.05 0.05P. roqueforti 0.15 0.01 0.05 0.05 0.05

pH 6.0 0.90 aw E. herbariorum 0.3 0.125 0.4 0.2 0.15E.rubrum 0.3 0.075 0.425 0.1 0.25A. flavus 0.3 0.075 0.3 0.05 0.15A. niger 0.55 0.075 0.3 0.05 0.1P. roqueforti 0.4 0.05 0.4 0.05 0.2

pH 5.0 0.85 aw E. herbariorum 0.05 0.05 0.05 0.1 0.05E.rubrum 0.05 0.01 0.05 0.05 0.05A. flavus 0.01 0.01 0.01 0.01 0.01A. niger 0.05 0.01 0.05 0.01 0.01P. roqueforti 0.01 0.01 0.01 0.01 0.01

pH 5.0 0.90 aw E. herbariorum 0.05 0.05 0.1 > 0.15 0.05E.rubrum 0.05 0.05 0.1 > 0.15 0.05A. flavus 0.05 0.05 0.1 0.01 0.05A. niger 0.1 0.05 0.1 0.05 0.05P. roqueforti 0.1 0.05 0.1 0.05 0.05

Y. Huang et al. / Food Microbiology 27 (2010) 33–36 35

3. Results & discussion

The MIC values determined in terms of the total amount of acidon a wt./wt. basis were species specific (Table 2). The MIC (% totalwt.) values for Eurotium species were generally higher than or equalto those for Aspergillus and Penicillium species, for each of the weakacids. The MIC values (% total wt.) for each species were lower at0.85 aw than at 0.90 aw for the same pH. This is most likely to be to

Table 3Estimated aqueous concentration (mmol/l) of total (MICt) and undissociated forms of we

Species aw/pH Estimated concentration (mmol/L) of weak acids in aq

sorbic dehydroacetic

MICt MICu MICt MICu

E. herbariorum0.85 aw pH 6.0 24.8 1.3 10.9 1.70.85 aw pH 5.0 8.3 3.0 5.6 3.50.90 aw pH 6.0 40.2 2.3 11.1 1.80.90 aw pH 5.0 6.8 2.4 4.5 2.9

E. rubrum0.85 aw pH 6.0 17.8 1.0 4.5 0.70.85 aw pH 5.0 4.5 1.6 0.6 0.40.90 aw pH 6.0 26.8 1.5 4.5 0.70.90 aw pH 5.0 4.5 1.6 3.0 1.9

A. flavus0.85 aw pH 6.0 8.9 0.5 3.0 0.50.85 aw pH 5.0 0.9 0.3 0.6 0.40.90 aw pH 6.0 26.8 1.5 4.5 0.70.90 aw pH 5.0 4.5 1.6 3.0 1.9

A. niger0.85 aw pH 6.0 26.8 1.5 3.0 0.50.85 aw pH 5.0 4.5 1.6 0.6 0.40.90 aw pH 6.0 49.1 2.7 4.5 0.70.90 aw pH 5.0 8.9 3.3 3.0 1.9

P. roqueforti0.85 aw pH 6.0 13.4 0.7 0.6 0.10.85 aw pH 5.0 0.9 0.3 0.6 0.40.90 aw pH 6.0 35.7 1.9 3.0 0.50.90 aw pH 5.0 8.9 3.3 3.0 1.9

the result of increased water stress at the lower aw. The MIC (% totalwt.) for capric acid, caprylic acid, and dehydroacetic acid werelower than for sorbic acid at pH 6.0, but at pH 5.0 the MIC (% totalwt.) of only caprylic acid and dehydroacetic acid remained lowerthan or equivalent to that of sorbic acid. These results indicate thatthe antifungal effect of caprylic acid and dehydroacetic acid, interms of MIC (% total wt.), is stronger than that of sorbic acid irre-spective of pH and aw within the range examined.

Woolford (1975) reported that the antifungal activity of caprylicacid > capric acid > caproic acid for six fungal species, including A.niger, at pH 5 and pH 6. Our results are similar, in that caproic acidhad the highest MIC (mmol/l) among the three acids either on a wt./wt. basis (Table 2) or on a molar basis (Table 3). However, theantifungal efficacies of caprylic acid and capric acid were depen-dent on the fungal species and pH and aw conditions.

The MIC values (% total wt.) for each fungal species were lowerat pH 5.0 than at pH 6.0 at equivalent aw values. Presumably thatobservation can be attributed to the higher proportion of undis-sociated form of each weak acid at pH 5.0 (Table 3), and thatundissociated form has been found to be largely responsible for theantimicrobial activities of weak acids (Stopforth et al., 2005). If theantimicrobial activity of a weak acid is related to the amount ofundissociated acid only, however, the MICu for each acid shouldremain the same at different pH levels. Eklund (1983) observed thatthat was not the case for sorbic acid. Our results for sorbic acid alsoshow that the MICu for sorbic acid is not the same at pH 5.0 and pH6.0 at equivalent aw values (Table 3). These results confirm theproposition by Eklund (1983) that the antimicrobial activity ofsorbic acid is related not only to its undissociated form but also toits dissociated form. Our results for the four weak acids also showthat the MICu for each acid was not the same at pH 5.0 and pH 6.0 atequivalent aw values. The undissociated portion (MICu) of all acidsat pH 6.0 was significantly lower than at pH 5.0 at both aw valuesemployed with the exception of the MICu of dehydroacetic acid at

ak acids (MICu) at the determined MIC.

ueous phase at their MIC

caproic capric caprylic

MICt MICu MICt MICu MICt MICu

31.9 2.2 8.1 0.6 12.8 0.98.0 3.5 10.7 4.6 6.5 2.8

51.7 3.6 17.4 1.2 15.6 1.112.9 5.6 >13.1 > 5.7 5.3 2.3

25.8 1.8 4.4 0.3 10.4 0.74.3 1.9 2.9 1.3 3.5 1.5

36.6 2.6 5.8 0.4 17.3 1.28.6 3.7 > 8.7 > 3.8 3.5 1.5

4.3 0.3 2.9 0.2 3.5 0.20.9 0.4 0.6 0.3 0.7 0.3

25.8 1.8 2.9 0.2 10.4 0.88.6 3.7 0.6 0.3 3.5 1.5

8.6 0.6 2.9 0.2 3.5 0.24.3 1.9 0.6 0.3 0.7 0.3

25.8 1.8 2.9 0.2 6.9 0.58.6 3.7 2.9 1.3 3.5 1.5

4.3 0.3 2.9 0.2 3.5 0.20.9 0.4 0.6 0.3 0.7 0.3

34.4 2.4 2.9 0.2 13.9 1.08.6 3.7 2.9 1.3 3.5 1.5

Y. Huang et al. / Food Microbiology 27 (2010) 33–3636

0.85 aw for E. rubrum, A. flavus, and A. niger (Table 3). These resultssuggest that both the dissociated and undissociated forms of thefour acids contributed to their antimicrobial activities under each ofthe pH and aw conditions employed.

Our results indicate that the inhibitory effect of weak acids onfungal growth is influenced by pH and aw and that it is specific tothe fungal species employed. Our work therefore extends, forexample the findings of Suhr and Nielsen (2004) who reported thataw levels and pH values are of paramount importance for the effi-cacy of the more commonly used propionate, sorbate, and benzoatein bakery products.

The four tested weak acids are listed as food additives in theUSA, with concentration limit of use specified for caproic acid (C6),caprylic acid (C8), and dehydroacetic acid, that is, 1, 1 and 0.0065%respectively (FDA, 2009). Our work indicates that only dehydro-acetic acid may be unsuitable for use in low-acid intermediatemoisture foods because the MIC of dehydroacetic acid was gener-ally over 0.01%. Caproic acid (C6), caprylic acid (C8) and capric acidshow potential for use in low-acid intermediate moisture food,especially caprylic acid, whose MIC of caprylic acid was lower than1% for all five fungal species under a wide range of low-acidintermediate moisture conditions.

References

Doores, S., 2005. Organic acids. In: Davidson, P.M., Sofos, J.N., Branen, A.L. (Eds.),Antimicrobials in Food, third ed. Taylor & Francis, Boca Raton FL, pp. 91–142.

Eklund, T., 1983. The antimicrobial effect of dissociated and undissociated sorbicacid at different pH levels. J. Appl. Bacteriol. 54, 383–389.

FDA, 2009. Listing of food additive status. http://www.fda.gov/Food/FoodIngredientsPackaging/FoodAdditives/FoodAdditiveListings/ucm091048.htm accessed 01.06.09.

Guynot, M.E., Ramos, A.J., Sanchis, V., Marin, S., 2005. Study of benzoate, propionate,and sorbate salts as mould spoilage inhibitors on intermediate moisture bakeryproducts of low pH (4.5–5.5). Int. J. Food Microbiol. 101, 161–168.

Kabara, J.J., Swieczkowski, D.M., Conley, A.J., Truant, J.P., 1972. Fatty acids andderivatives as antimicrobial agents. Antimicrob. Agents Chemother. 2, 23–28.

Kabara, J.J., Marshall, D.L., 2005. Medium-chain fatty acids and esters. In:Davidson, P.M., Sofos, J.N., Branen, A.L. (Eds.), Antimicrobials in Food, third ed.Taylor & Francis, Boca Raton, FL, pp. 327–360.

Kato, N., Shibasaki, I., 1975. Comparison of antimicrobial activities of fatty acids andtheir esters. Hakko kogaku zasshi ¼ J. Ferment. Technol. 53, 793–801.

Lide, D.R., Frederikse, H.P.R., 1997. CRC Handbook of Chemistry and Physics. CRCPress, Boca Raton, New York, pp. 8–45 to 8–55.

Malfeito Ferreira, M., Loureiro-Dias, M.C., Loureiro, V., 1997. Weak acid inhibition offermentation by Zygosaccharomyces bailii and Saccharomyces cerevisiae. Int.J. Food Microbiol. 36, 145–153.

Morozumi, S., Wauke, T., Hitokoto, H., 1985. Effects of preservatives on the growthand mycotoxin production of toxinogenic fungi. Ann. Rep. Tokyo Metr Res. Lab.Public Health 36, 143–150.

Mukaiyama, T., Ozake, S., Hoshino, T., 1954. On the thermal dissociation of organiccompounds. VI. The effect of the substituent and that of the solvent on thethermal dissociation of ureas. Bull. Chem. Soc. Jpn. 27, 578–582.

Pitt, J.I., Hocking, A.D., 1997. Fungi and Food Spoilage, second ed.. Blackie Academic& Professional, London, pp. 509–510.

Ray, P., Liewen, M.B., 2004. Antifungal food additives. In: Arora, D.K. (Ed.), FungalBiotechnology in Agricultural, Food, and Environmental Applications. MarcelDekker, New York, pp. 291–297.

Rihakova, Z., Filip, V., Smidrkal, J., 2001. Antifungal effectiveness of glycerol esterswith lauric acid and its dependency on their purity. Adv. Food Sci. 23, 57–62.

Skirdal, I.M., Eklund, T., 1993. Microculture model studies on the effect of sorbic acidon Penicillium chrysogenum, Cladosporium cladosporidoides and Ulocladiumatrum at different pH levels. J. Appl. Bacteriol. 74, 191–195.

Skrivanova, E., Marounek, M., Dlouha, G., Kanka, J., 2005. Susceptibility ofClostridium perfringens to C2-C18 fatty acids. Lett. Appl. Microbiol. 41, 77–81.

Stopforth, J.D., Sofos, J.N., Busta, F.F., 2005. Sorbic acid and sorbates. In:Davidson, P.M., Sofos, J.N., Branen, A.L. (Eds.), Antimicrobials in Food, third ed.Taylor & Francis, Boca Raton FL, pp. 49–90.

Suhr, K.I., Nielsen, P.V., 2004. Effect of weak acid preservatives on growth of bakeryproduct spoilage fungi at different water activities and pH values. Int. J. FoodMicrobiol. 95, 67–78.

Woolford, M.K., 1975. Microbiological screening of the straight chain fatty acids(C1-C12) as potential silage additives. J. Sci. Food Agric. 26, 219–228.