Characterization of Volatile Compounds in Microfiltered Pasteurized Milk Using Solid-Phase Microextraction and GC×GC-TOFMS

Abstract

Volatiles of milk were characterized by solid-phase microextraction coupled with comprehensive two-dimensional gas chromatography and time-of-flight mass spectrometry. Comprehensive two-dimensional gas chromatography with time-of-flight mass spectrometry two-dimensional gas chromatography with time-of-flight mass spectrometry not only separated the 52 compounds that co-elute in conventional gas chromatography-mass spectrometry, but also identified 107 compounds that were first reported in milk. These volatiles included aliphatic hydrocarbons (69), aromatic hydrocarbons (42), ketones (28), esters (16), aldehydes (14), alcohols (14), acids (14), nitrogenous compounds (9), ethers (8), and sulfo compounds (3). Five dominant volatiles were hexanoic acid (193.57 ng/mL milk), methoxy-phenyl-oxime (114.83 ng/mL milk), octanoic acid (109.38 ng/mL milk), 4,5-dimethyl-1-hexene (101.48 ng/mL milk), and 2-Pentanone (99.74 ng/mL milk). This proposes an improvement methodology for determining the volatiles of dairy products.

Keywords:

• Milk
• Volatile compounds
• GC×GC-TOFMS
• Headspace solid-phase microextraction (HS-SPME)

INTRODUCTION

Milk flavor is one of the most important factors that determine the consumer acceptability and its shelf life. Fresh milk flavor is delicate and subtle and can be overshadowed by off-flavor, which reduce the sensory quality and economic value of dairy products.^[1,2] The production of volatiles in milk is influenced by several factors: cow species, daily diets, environment, age, biological factors, processing, and storage conditions.^[3–9] Some distinctive volatiles have been used to identify the freshness, deterioration, and predict shelf life of milk.^[10,11] Therefore, the qualitative and quantitative measurements of such volatiles have attracted much interest in recent years.

A common analytical technique identifying volatiles in milk is gas chromatography-mass spectrometry (GC-MS) analysis. Twenty-four volatile organic compounds related to flavor were detected during milk fermentation by Lactobacillus pentosus using headspace solid-phase microextraction (HS-SPME)/GC-MS.^[9] A previous study showed that 37 volatiles including aldehydes, ketones, fatty acids, esters, alcohols, and aromatic compounds were identified in the headspace of milk samples by GC-MS.^[7] A total of 59 volatile components were determined by GC-MS in skim milk and non-standardized milk pasteurized at 80, 100, and 120°C.^[12] Chouliara et al. employed SPME-GC-MS to analyze non-sonicated pasteurized milk and 17 compounds were semi-quantified.^[6] Solano-Lopez et al. applied GC-MS to identify 16, 22, and 26 compounds in ultrapasteurized-polyethylene terephthalate (PET)-bottled milk samples stored for 1, 30, and 60 days, respectively.^[3] Perkins et al. detected 17 stale flavor volatiles in ultrahigh-temperature (UHT)-processed milk using SPME-GS-MS.^[13] Nineteen volatile organic compounds were isolated and identified using HS-SPME-GC-mass spectrometry/flame ionization detection (MS/FID) in milk tainted with off-flavor.^[14] Contarini et al. observed 11 volatile compounds in both whole and partially skimmed UHT milk using GC-MS.^[15]

However, GC-MS may not be sufficient to separate all the volatile components, because it could result in co-elutions of several components. Two-dimensional gas chromatography with time-of-flight mass spectrometric detection (GC×GC-TOFMS) is a suitable alternative for qualitative and quantitative analysis of complex matrices, such as food and beverage volatiles, due to its increased peak capacity, sensitivity, and selectivity.^[16–19] Structurally organized distributions of different classes of compounds are another advantage of GC×GC-TOFMS, which assists the identification of unknown compounds.^[18] Orthogonal 2D separation of lavender essential oil led to a 25-fold increase in sensitivity and a 3-fold increase in the number of resolved components compared with GC-MS analysis.^[16]

The detailed profiling of volatiles from milk is informative, not only to classify and assess dairy products, but also to predict the shelf life of dairy products on the basis of sensory profiles (aroma, off-flavor, and taste).^[20] However, to the best of the authors’ knowledge, the volatiles of milk and other dairy products have not yet been studied using GC×GC-TOFMS. Thus, the aim of this work was to develop a methodology based on HS-SPME coupled with GC×GC-TOFMS to obtain a qualitative and quantitative characterization of the total volatile components of milk. One-dimensional GC-MS detection was also used as a comparative method.

MATERIALS AND METHODS

Materials

Fresh raw milk (3.4% fat), collected from the Jinshan dairy farm (Shanghai, China), was centrifuged to at 4000 g for 30 min to remove the fat and leucocytes. After centrifugation, the skim milk (less than 0.1% fat) was microfiltrated through ceramic membranes (Tami, France; nominal pore size:1.4 μm; total membrane surface area:13.3 m²; flow rate: 5t·h^–1; inlet pressure: 2.5 bar) at 50°C. The separated cream stream (35% fat) of the fresh milk was heated at 120°C for 4 s and then mixed with the permeate of microfiltration to the fat-adjusted whole milk (3.4% fat). Then the mixture was homogenized (150/50 bar) at 65°C and pasteurized at 72°C for 15 s, and cooled to 4°C and then filled to give a stable product for volatiles analysis. All of the volatile standards used for identification and other reagents were of GC-analytical grade and were obtained from Sigma-Aldrich. All other chemicals were purchased from Sinopharm Chemical Reagent Co., Ltd, China.

Solid Phase Microextraction

Milk volatiles were collected by SPME technique in an auto sampler (GC Sampler 80, Agilent, Santa Clara, CA, USA), following the previous method^[7] with some modifications. A milk sample of 10 mL, 2 μL of the internal standard solution (0.1 mg.mL⁻¹ 2,4,6-trimethylpyridine in methyl alcohol) and 3.7 g of sodium chloride were placed in an annealed 20-mL brown glass vial with screw cap. The SPME fiber used was 50/30 μm DVB/CAR on PDMS (Supelco, Bellefonte, PA, USA). The mixtures were incubated at 50°C for 20 min with 500 rpm vibration and the SPME fiber was exposed to the headspace above the milk sample for 30 min at 50°C with 250 rpm vibration to allow adsorption of volatiles. After extraction the volatile compounds were desorbed from the fiber coating by inserting the fiber into the GC injector for analysis.

GC-MS Analysis

Headspace of the volatile compounds was analyzed using a 7890A gas chromatograph coupled with a 5975C mass spectrometer (Agilent Technologies, Palo Alto, CA, USA). A DB-5MS column (30 m × 0.25 mm i.d., 0.25 µm df) from Agilent was used to separate the compounds. The injector port was heated to 260°C and injections (1 μL) were performed in splitless mode. Helium (purity >99.999%) was used as the carrier gas at a constant flow of 1 mL min^–1. The column temperature was held at 40°C for 5 min, increased to 220°C at 5°C min^–1, and then to 250°C at 20°C min^–1 and held for 7.5 min until the end of the program. The temperatures of the ion source and interface were maintained at 230 and 280°C, respectively. Electron impact ionization mass spectra were recorded with an ionization energy of 70 eV and an EM voltage of 1847 V. Mass spectra were scanned from 20 to 350 amu in total ion chromatogram (TIC) mode to identify the compounds based on the Saturn spectra reported on the National Institute of Standards and Technology (NIST) 2011 Mass Spectral Library (MainLib: 212,961 spectra and Replib: 30,932 spectra). Main, molecular, and qualifier ions were selected for each compound identified. Relative retention times of detected compounds were also determined by injecting 20 μL of 0.1 mg.mL^–1 n-alkane standard solutions (ASTM D2887-01 Calibration Mix, Restek, Pennsylvania, USA) with a split ratio of 1:200. Signals were processed using Agilent MSD Productivity ChemStation Enhanced Data Analysis software (Agilent, Santa Clara, CA, USA).

GC×GC-TOFMS Analysis

The GC×GC-TOFMS system used was a gas chromatograph 6890N Agilent (Agilent Technologies) with a LECO Pegasus 4D TOFMS system (St. Joseph, MI, USA). The first dimensional chromatographic separation column was a DB-5MS (30 m × 0.25 mm i.d., 0.25 µm df). The second dimension column was DB-17HT (2 m × 0.1 mm i.d., 0.15 μM df). Highly purified helium (99.999%) was selected as the carrier gas, with a flow rate of 1.0 mL min^–1. Samples were incubated at 50°C for 20 min and the fiber was exposed to the vial head space during 30 min at 50°C. The adsorbed volatiles were desorbed in the GC×GC injector port in splitless mode at 260°C for 1 min. All samples (1 µL) were introduced into the GC×GC inlet system by an auto sampler using splitless mode injection at 260°C. The primary oven temperature program was initially set at 40°C for 5 min, increased to 220°C at a rate of 5°C.min^–1, further increased to 250°C at a rate of 20°C min^–1 and maintained for 7.5 min. The secondary oven temperature was 5°C higher than the primary one. The modulator temperature was 15°C higher than the second oven temperature, and the modulation period was 4 s. The TOFMS was operated in the electronic impact ionization mode at 70 eV at an acquisition rate of 100 spectra s^–1 and a mass range of m/z 20–350 amu in TIC mode. The ion source temperature was 220°C, the interface temperature was 270°C, and the detector was operated at 1700 V. Mass spectral library matching was carried out by using an NIST Mass Spectral Library (MainLib: 212,961 spectra and Replib: 30,932 spectra).

Data Acquisition and Quantification

In this study, the data were processed by ChromaTOF software (version 4.4, Leco, St. Joseph, MI, USA). Matching its deconvoluted mass spectra with the NIST Mass Spectral Library (MainLib: 212,961 spectra and Replib: 30,932 spectra) identified each individual peak. Only peaks with a mass spectral match factor (similarity) higher than 800 were regarded as correctly identified and retained for further grouping and analysis. Similarity ranging from 850 to 969 and the quality of the mass spectrum of each compound were manually inspected. The linear retention indices (RIs) in the GC-MS and first dimension of GC×GC-TOFMS were calculated by interpolation of the retention times of a C7-C40 n-alkanes series under the same chromatographic conditions, by the ChromaTOF software system. The calculated RIs were compared with those in the NIST Mass Search 2.0 and other references.^[10,21–25]

RESULTS AND DISCUSSION

SPME-GC-MS Analysis

The volatile profile of milk obtained by SPME-GC-MS is shown in Tables 1 and 2. In total, 52 identified compounds were organized in following chemical groups: aliphatic hydrocarbons, alcohols, aldehydes, ketones, acids, esters, aromatic hydrocarbons, nitrogenous compounds, ethers, and sulfo compounds. Both the acids (11) and aromatic hydrocarbons (11) were the predominant groups among the identified components. The top four acid compounds were octanoic acid (66.94 ng/mL milk), hexanoic acid (47.67 ng/mL milk), n-Decanoic acid (42.56 ng/mL milk), and butanoic acid (11.17 ng/mL milk), and the top four aromatic hydrocarbons were toluene (10.54 ng/mL milk), phenol (9.34 ng/mL milk), ethylbenzene (5.15 ng/mL milk), and styrene (2.82 ng/mL milk; Table 1).

TABLE 1 Milk volatile compounds identified using GC×GC-TOF MS and GC-MS

			GC×GC-TOF MS					GC-MS				Reported
	Compounds	CAS number	^1tR (s)	^2tR (s)	RI-cal^a	RI-ref^b	C (ng/mL milk)	tR (s)	RI-cal^c	RI-ref^d	C(ng/mLmilk)	Threshold
Aliphatic Hydrocarbons
1	3-Hexene, (E)-	13269-52-8	320	1.8	626	604	8.74 ± 0.48	–	–	–	–
2	Pentane, 2-methyl-	107-83-5	328	1.8	630	596	6.41 ± 0.11	–	–	–	–
3	Pentane, 3-methyl-	96-14-0	340	1.9	635	589	25.04 ± 1.77	–	–	–	–
4	1-Pentene, 2-methyl-	763-29-1	348	1.8	638	588	13.76 ± 1.82	–	–	–	–
5	Cyclopentane, methyl-	96-37-7	396	1.9	658	638	15.82 ± 0.66	–	–	–	–
6	1-Hexene, 4,5-dimethyl-	16106-59-5	432	1.9	674	740	101.48 ± 2.25	–	–	–	–
7	Cyclohexane	110-82-7	444	2	679	669	4.15 ± 0.36	–	–	–	–
8	Hexane, 3-methyl-	589-34-4	460	1.9	686	673	0.87 ± 0.02	–	–	–	–
9	2-Hexene, 5-methyl-, (E)-	7385-82-2	488	2	698	669	10.27 ± 0.77	–	–	–	–
10	Heptane	142-82-5	504	1.9	705	700	9.61 ± 0.65	–	–	–	–
11	Cyclohexane, methyl-	108-87-2	560	2.1	729	720	2.65 ± 0.42	–	–	–	–
12	Hexane, 3-methyl-4-methylene-	3404-67-9	628	2	758	756	20.69 ± 0.72	–	–	–	–
13	Heptane, 3-methylene-	1632-16-2	696	2.1	787	791	30.10 ± 1.65	–	–	–	–
14	Octane	111-65-9	720	2	798	800	6.29 ± 0.65	–	–	–	–
15	3-Octene, (Z)-	14850-22-7	720	2.1	798	800	11.55 ± 0.10	–	–	–	–
16	2-Octene, (E)-	13389-42-9	736	2.1	808	811	3.75 ± 0.40	–	–	–	–
17	Cyclopentane, propyl-	2040-96-2	800	2.1	832	833	3.01 ± 0.02	–	–	–	–
18	Butane, 2-cyclopropyl-	5750-02-7	836	2.1	848	813	2.07 ± 0.15	–	–	–	–
19	Octane, 4-methyl-	2216-34-4	872	2	864	849	0.70 ± 0.07	–	–	–	–
20	Heptane, 3-ethyl-	15869-80-4	884	2	868	872	2.02 ± 0.15	–	–	–	–
21	4-Nonene	2198-23-4	948	2.1	895	887	3.14 ± 0.15	–	–	–	–
22	Nonane	111-84-2	956	2	900	900	11.67 ± 2.62	–	–	–	–
23	2-Pentene, 3-methyl-	922-61-2	1032	2.1	934	936	7.21 ± 0.14	–	–	–	–
24	3-Octene, 2,6-dimethyl-	6874-28-8	1048	2.1	942	895	26.83 ± 3.48	–	–	–	–
25	1,6-Octadiene, 3,7-dimethyl-, (S)-	10281-55-7	1052	2.1	943	945	9.16 ± 0.16	–	–	–	–
26	Octane, 2,3,3-trimethyl-	62016-30-2	1064	2	949	966	5.35 ± 0.35	–	–	–	–
27	Octane, 4-ethyl-	15869-86-0	1076	2	954	954	3.85 ± 0.35	–	–	–	–
28	Octane, 2,2,6-trimethyl-	62016-28-8	1088	2	959	963	0.69 ± 0.21	–	–	–	–
29	2-Octene, 3,7-dimethyl-, (Z)-	6874-32-4	1104	2.1	967	970	2.96 ± 0.33	–	–	–	–
30	Cyclohexane, 1-ethyl-4-methyl-, cis-	4926-78-7	1112	2.3	971	941	0.46 ± 0.07	–	–	–	–
31	Beta-pinene	127-91-3	1136	2.3	982	981	14.14 ± 2.71	–	–	–	–
32	Nonane, 3-methylene-	51655-64-2	1144	2.1	985	987	3.82 ± 0.53	–	–	–	–
33	Heptane, 5-ethyl-2-methyl-	13475-78-0	1160	2	993	925	46.11 ± 6.45	–	–	–	–
34	Decane	124-18-5	1176	2	999	1050	7.13 ± 0.48	317	1000	1000	1.04 ± 0.18
35	1,1,3,3,5-Pentamethylcyclohexane	70810-19-4	1176	2.1	1000	1000	3.67 ± 1.21	–	–	–	–
36	Heptane, 2,2,4,6,6-pentamethyl-	13475-82-6	1220	2	1022	1032	13.23 ± 1.29	–	–	–	–
37	2-Octene, 4-ethyl-	74630-09-4	1220	2.1	1022	nf	6.73 ± 0.94	–	–	–	–
38	1-Decyne	764-93-2	1220	2.1	1022	1014	1.56 ± 0.05	–	–	–	–
39	Cyclohexene, 1-methyl-4-(1-methylethenyl)-, (S)-	5989-54-8	1240	2.3	1032	1020	3.04 ± 0.29	–	–	–	–
40	Octane, 2,3,6,7-tetramethyl-	52670-34-5	1248	2	1036	958	6.34 ± 2.15	–	–	–	–
41	1,3,8-p-Menthatriene	18368-95-1	1296	2.5	1060	1111	2.65 ± 0.07	–	–	–	–
42	Decane, 2-methyl-	6975-98-0	1340	2	1082	1062	22.14 ± 0.35	–	–	–	–
43	trans-4-Decene	19398-89-1	1360	2.1	1092	1023	18.60 ± 2.15	–	–	–	–
44	Undecane	1120-21-4	1376	2	1100	1100	5.44 ± 0.11	485	1096	1100	0.83 ± 0.08
45	Nonane, 3-methyl-	5911-04-6	1396	2	1111	1051	12.68 ± 2.51	–	–	–	–
46	Octane, 3-ethyl-2,7-dimethyl-	62183-55-5	1408	2	1118	1180	4.23 ± 0.23	–	–	–	–
47	Pentane, 2,3,3-trimethyl-	560-21-4	1416	2	1122	1073	3.42 ± 0.69	–	–	–	–
48	2,2,7,7-Tetramethyloctane	1071-31-4	1440	2	1134	1071	15.63 ± 0.31	–	–	–	–
49	Undecane, 4-methyl-	2980-69-0	1440	2	1134	1163	13.90 ± 0.18	–	–	–	–
50	Cyclohexane, pentyl-	4292-92-6	1452	2.2	1141	1129	2.08 ± 0.16	–	–	–	–
51	Undecane, 2-methyl-	7045-71-8	1496	2	1165	1169	7.25 ± 0.35	–	–	–	–
52	Undecane, 3-methyl-	1002-43-3	1508	2	1172	1177	0.52 ± 0.02	–	–	–	–
53	3-Undecene, 3-methyl-	23381-94-4	1532	2.1	1184	1199	2.43 ± 0.25	–	–	–	–
54	5-Dodecene, (E)-	7206-16-8	1544	2.1	1191	1189	1.19 ± 0.11	–	–	–	–
55	Cyclooctane, 1,4-dimethyl-, cis-	13151-99-0	1544	2.1	1191	nf	8.81 ± 0.35	–	–	–	–
56	Dodecane	112-40-3	1560	2	1200	1200	3.76 ± 0.66	693	1194	1200	1.44 ± 0.12
57	Undecane, 2,6-dimethyl-	17301-23-4	1584	2	1215	1217	1.20 ± 0.06	–	–	–	–
58	Decane, 2,3,5,8-tetramethyl-	192823-15-7	1644	2.1	1253	1318	0.95 ± 0.28	–	–	–	–
59	Dodecane, 2-methyl-	1560-97-0	1672	2.1	1271	1267	1.03 ± 0.15	–	–	–	–
60	2-Tridecene, (Z)-	41446-59-7	1716	2.1	1300	1296	0.96 ± 0.08	–	–	–	–
61	Tridecane	629-50-5	1732	2	1308	1300	0.83 ± 0.01	881	1298	1300	0.61 ± 0.09
62	Tridecane, 3-methyl-	6418-41-3	1848	2	1373	1367	6.05 ± 0.18	–	–	–	–
63	2-Tridecene, (E)-	41446-58-6	1880	2.1	1391	1321	2.58 ± 0.27	–	–	–	–
64	Tetradecane	629-59-4	1892	2	1398	1400	1.97 ± 0.17	1055	1397	1400	1.40 ± 0.42
65	Bicyclo[7.2.0]undec-4-ene, 4,11,11-trimethyl-8-methylene-,[1R-(1R*,4Z,9S*)]-	118-65-0	1948	2.5	1435	1419	1.14 ± 0.06	–	–	–	–
66	Tetradecane, 3-methyl-	18435-22-8	2000	2	1479	1472	0.8 ± 0.14	–	–	–	–
67	3-Tetradecene, (Z)-	41446-67-7	2032	2.1	1491	1489	0.86 ± 0.09	–	–	–	–
68	Pentadecane	629-62-9	2044	2	1500	1500	0.43 ± 0.19	–	–	–	–
69	Hexadecane	544-76-3	2188	2	1610	1600	17.48 ± 1.41	–	–	–	–
Alcohols
70	1,5-Hexadien-3-ol	924-41-4	476	2.4	693	760	56.13 ± 8.82	–	–	–	–	500^C
71	1-Butanol	71-36-3	556	2.4	694	663	0.87 ± 0.06	607	1156	1153	2.04 ± 0.10
72	Ethanol, 2-ethoxy-	110-80-5	564	2.5	731	702	11.33 ± 1.7	747	1126	nf	1.65 ± 0.26
73	3-Buten-1-ol, 3-methyl-	763-32-6	572	2.5	733	705	7.96 ± 0.1	–	–	–	–
74	2-Pentanol, 2-methyl-	590-36-3	576	2.3	736	731	18.07 ± 0.99	–	–	–	–
75	3-Pentanol, 3-methyl-	77-74-7	620	2.3	755	738	10.39 ± 0.64	–	–	–	–
76	1-Pentanol	71-41-0	648	2.4	767	735	18.64 ± 0.48	–	–	–	–	4000^B
77	2-Hexanol	626-93-7	724	2.4	800	793	4.48 ± 0.39	–	–	–	–
78	1-Hexanol	111-27-3	884	2.4	869	870	6.78 ± 0.32	–	–	–	–	500^C
79	1-Octen-3-ol	3391-86-4	1132	2.4	980	981	7.19 ± 1.48	–	–	–	–	0.01^A
80	2-Octanol, (S)-	6169-06-8	1176	2.3	1000	997	2.58 ± 0.27	–	–	–	–
81	1-Hexanol, 2-ethyl-	104-76-7	1232	2.4	1028	1031	2.35 ± 0.57	1205	1492	1492	0.73 ± 0.20
82	1-Octanol	111-87-5	1316	2.4	1070	1078	0.64 ± 0.06	1307	1560	1547	1.08 ± 0.08
83	2-Furanmethanol, 5-ethenyltetrahydro-a,a,5-trimethyl-, cis-	5989-33-3	1324	2.5	1074	1064	0.55 ± 0.08	–	–	–	–
Aldehydes
84	Propanal, 2-methyl-	78-84-2	320	2	626	600	2.79 ± 0.24	–	–	–	–
85	Butanal	123-72-8	352	2.2	640	600	8.77 ± 0.62	–	–	–	–	130^C
86	Butanal, 3-methyl- (off-flavor)	590-86-3	432	2.3	674	667	22.86 ± 0.12	–	–	–	–	0.25 ^A
87	Butanal, 2-methyl-	96-17-3	448	2.3	681	678	59.41 ± 1.49	–	–	–	–	0.9 ^A
88	Hexanal	66-25-1	724	2.5	799	802	4.23 ± 0.63	473	1089	1105	4.41 ± 0.79	10.5 ^A
89	Heptanal	111-71-7	960	2.5	901	899	17.11 ± 0.71	–	–	–	–	3 ^A
90	Octanal	124-13-0	1180	2.5	1002	1003	0.26 ± 0.02	870	1291	1286	0.88 ± 0.13	3^B
91	Nonanal	124-19-6	1384	2.4	1104	1108	1.12 ± 0.12	1050	1394	1383	0.91 ± 0.32	1^B
92	Decanal	112-31-2	1572	2.4	1207	1208	2.87 ± 0.15	–	–	–	–	2^B
93	2-Decenal, (Z)-	2497-25-8	1668	2.5	1302	1271	6.41 ± 0.09	–	–	–	–	0.4^B
94	Undecanal	112-44-7	1744	2.4	1315	1303	0.67 ± 0.09	–	–	–	–
95	2,4-Decadienal	2363-88-4	1768	2.6	1328	1325	0.27 ± 0.00	–	–	–	–	0.07^B
96	2-Undecenal, E-	53448-07-0	1840	2.5	1368	1360	1.32 ± 0.29	–	–	–	–
97	Dodecanal	112-54-9	1908	2.4	1408	1411	0.55 ± 0.07	–	–	–	–
Ketones
98	Acetone	67-64-1	332	1.9	610	515	29.40 ± 5.04	149	863	806	87.44 ± 12.3
99	2-Butanone	78-93-3	356	2.4	642	606	71.10 ± 2.96	200	905	906	32.55 ± 4.62
100	2-Butanone, 3-methyl-	563-80-4	436	2.4	676	707	38.98 ± 2.69	–	–	–	–
101	3-Buten-2-one, 3-methyl-	814-78-8	460	2.4	685	693	4.05 ± 0.03	–	–	–	–	500^C
102	2-Pentanone	107-87-9	480	2.5	695	701	99.74 ± 7.76	287	976	962	4.33 ± 1.40
103	4-Methyl-2-pentanone	108-10-1	584	2.4	739	740	16.0 ± 0.85	–	–	–	–
104	2-Pentanone, 3-methyl-	565-61-7	612	2.5	751	753	4.34 ± 0.83	–	–	–	–
105	2-Hexanone	591-78-6	696	2.5	787	789	89.07 ± 2.97	–	–	–	–	200^C
106	2-Pentanone, 4-hydroxy-4-methyl-	123-42-2	828	2.8	843	834	2.14 ± 0.60	–	–	–	–
107	(R)-(+)-3-Methylcyclopentanone	6672-30-6	840	2.9	850	848	1.20 ± 0.07	–	–	–	–
108	Cyclopentanone	120-92-3	840	3.1	850	851	1.82 ± 0.01	–	–	–	–
109	3-Heptanone	106-35-4	920	2.5	884	866	7.86 ± 0.7	–	–	–	–
110	2-Heptanone	110-43-0	928	2.5	888	893	60.47 ± 1.01	669	1185	1136	22.19 ± 3.01
111	Cyclohexanone	108-94-1	948	3.1	896	898	4.83 ± 0.19	–	–	–	–
112	Butyrolactone	96-48-0	992	0.1	914	891	0.91 ± 0.03	1415	1635	1635	1.98 ± 0.09
113	2,5-Hexanedione	110-13-4	1016	3.3	926	930	0.32 ± 0.03	1230	1508	1505	0.41 ± 0.07
114	3-Octanone	106-68-3	1140	2.5	983	988	7.40 ± 0.19	–	–	–	–
115	5-Hepten-2-one, 6-methyl-	110-93-0	1140	2.6	983	987	1.06 ± 0.05	–	–	–	–
116	2-Octanone	111-13-7	1152	2.5	989	994	3.46 ± 0.59	–	–	–	–
117	Acetophenone	98-86-2	1316	3.2	1070	1052	0.97 ± 0.04	–	–	–	–
118	2-Nonanone	821-55-6	1356	2.4	1090	1093	12.03 ± 0.71	1042	1390	1380	7.25 ± 1.07	200^B
119	2H-Pyran-2-one, tetrahydro-6-methyl-	823-22-3	1364	3.7	1094	1056	1.42 ± 0.19	–	–	–	–
120	(1R-cis)-, Ethanone, 1-(1,2,2,3-tetramethylcyclopentyl)-	59642-07-8	1428	2.1	1128	1165	15.64 ± 3.8	–	–	–	–
121	1-Phenyl-2-butanone	1007-32-5	1612	3.1	1234	1240	6.50 ± 0.11	–	–	–	–
122	2H-Pyran-2-one, tetrahydro-6-propyl-	698-76-0	1708	3.4	1295	1270	0.23 ± 0.01	1854	1974	1983	0.30 ± 0.04
123	2-Undecanone	112-12-9	1720	2.4	1302	1296	1.91 ± 0.41	1366	1560	1585	3.55 ± 0.21	500^C
124	2H-Pyran-2-one, tetrahydro-6-pentyl-	705-86-2	2040	3.2	1716	1786	0.77 ± 0.03	2114	2184	nf	1.67 ± 0.17
125	2-Pentadecanone, 6,10,14-trimethyl-	502-69-2	2496	2.1	1851	1823	0.26 ± 0.06	–	–	–	–
Acids
126	Acetic acid	64-19-7	400	2.7	658	660	1.97 ± 0.33	1168	1468	1455	0.51 ± 0.08	22000 ^A
127	L-Lactic acid	79-33-4	556	2.7	727	838	4.26 ± 0.23	–	–	–	–
128	Butanoic acid	107-92-6	892	2.6	873	856	62.50 ± 4.52	1424	1751	1620	11.17 ± 1.50
129	Hexanoic acid	142-62-1	1132	2.6	980	973	193.57 ± 23.2	1712	1860	1829	47.67 ± 4.24
130	Pentanoic acid	109-52-4	1136	2.5	981	934	3.87 ± 0.51	–	–	–	–
131	Heptanoic acid	111-14-8	1304	2.5	1064	1074	3.34 ± 0.49	1845	2066	1954	0.64 ± 0.08
132	Octanoic acid	124-07-2	1616	2.5	1167	1156	109.38 ± 2.25	1970	2068	2044	66.94 ± 4.40
133	Nonanoic acid	112-05-0	1660	2.4	1264	1263	1.36 ± 0.10	2091	2165	2149	0.70 ± 0.15
134	9-Decenoic acid	14436-32-9	1812	2.5	1353	1389	1.84 ± 0.23	2270	2310	2305	2.22 ± 0.33
135	n-Decanoic acid	334-48-5	1824	2.5	1359	1367	38.57 ± 3.22	2205	2257	2263	42.56 ± 4.70
136	Undecanoic acid	112-37-8	1976	2.4	1559	1556	3.55 ± 0.62	–	–	–	–
137	Dodecanoic acid	143-07-7	2120	2.4	1600	1553	1.73 ± 0.72	2420	2431	2471	4.23 ± 0.32
138	Tetradecanoic acid	544-63-8	2388	2.5	1754	1748	4.19 ± 0.82	2577	>2500	2692	1.35 ± 0.42
139	n-Hexadecanoic acid	57-10-3	2600	2.1	1919	1942	2.11 ± 0.24	2717	>2500	2886	1.72 ± 1.42
Esters
140	Ethyl Acetate	141-78-6	376	2.2	648	634	39.33 ± 7.25	190	897	890	16.6 ±1.20	5
141	n-Propyl acetate	109-60-4	532	2.4	717	716	14.95 ± 1.77	–	–	–	–
142	Butanoic acid, methyl ester	623-42-7	552	2.4	724	696	8.53 ± 1.04	303	989	nf	1.33 ± 0.21
143	sec-Butyl acetate	105-46-4	624	2.4	757	746	8.60 ± 0.46	–	–	–	–
144	Isobutyl acetate	110-19-0	660	2.4	774	767	0.17 ± 0.03	–	–	–	–
145	Acetic acid, butyl ester	123-86-4	752	2.5	812	791	14.58 ± 1.30	458	1081	1041	1.65 ± 0.21
146	1-Methoxy-2-propyl acetate	108-65-6	872	2.7	864	857	1.07 ± 0.17	–	–	–	–
147	Propanoic acid, butyl ester	590-01-2	972	2.4	907	888	0.88 ± 0.07	589	1146	nf	1.73 ± 0.25
148	Hexanoic acid, methyl ester	106-70-7	1008	2.4	923	904	1.00 ± 0.21	–	–	–	–
149	Butanoic acid, butyl ester	109-21-7	1160	2.4	992	978	0.33 ± 0.05	739	1221	nf	0.36 ± 0.22
150	Hexanoic acid, ethyl ester	123-66-0	1168	2.4	996	1017	3.5 ± 0.31	–	–	–	–
151	Dihydro-2(3H)-thiophenone	1003-10-7	1180	3.8	1002	952	0.6 ± 0.02	–	–	–	–
152	Octanoic acid, methyl ester	111-11-5	1416	2.4	1121	1108	0.96 ± 0.07	–	–	–	–
153	3-Methylheptyl acetate	72218-58-7	1460	2.3	1145	nf	0.67 ± 0.09	–	–	–	–
154	Decanoic acid, methyl ester	110-42-9	1768	2.4	1329	1309	0.43 ± 0.11	–	–	–	–
155	Propanoic acid, 2-methyl-, 1-(1,1-dimethylethyl)-2-methyl-1,3-propanediyl ester	74381-40-1	2172	2.5	1600	1605	1.43 ± 0.09	–	–	–	–
Aromatic Hydrocarbons
156	Benzene	71-43-2	440	2.4	678	677	14.10 ± 3.28	–	–	–	–
157	Toluene	108-88-3	648	2.5	767	773	23.51 ± 4.15	392	1043	1028	10.54 ± 0.61	1000^B
158	Ethylbenzene	100-41-4	868	2.6	862	868	10.26 ± 2.73	550	1127	1118	5.16 ± 0.60
159	Benzene, 1,3-dimethyl-	108-38-3	888	2.5	882	872	49.34 ± 11.83	–	–	–	–
160	Styrene	100-42-5	940	2.7	893	914	6.46 ± 0.93	812	1260	1265	2.82 ± 0.35
161	Benzene, propyl-	103-65-1	1080	2.5	956	959	1.25 ± 0.12	–	–	–	–
162	Benzaldehyde	100-52-7	1100	3.2	965	965	2.12 ± 0.28	1259	1528	1531	2.38 ± 0.21	350^B
163	Benzene, 1-ethyl-2-methyl-	611-14-3	1112	2.6	971	972	10.75 ± 0.50	–	–	–	–
164	Benzene, 1,2,3-trimethyl-	526-73-8	1112	2.5	971	998	1.21 ± 0.12	850	1281	1325	0.78 ± 0.15
165	Phenol	108-95-2	1124	3	975	995	13.84 ± 0.19	1908	2017	2015	9.34 ± 4.74
166	Benzonitrile	100-47-0	1144	3.3	986	973	1.69 ± 0.38	–	–	–	–
167	o-Cymene	527-84-4	1232	2.5	1028	1034	0.84 ± 0.07	830	1269	1253	1.50 ± 0.15
168	Indane	496-11-7	1256	2.8	1040	1055	0.43 ± 0.09	–	–	–	–
169	Benzeneacetaldehyde	122-78-1	1272	3.2	1047	1061	0.38 ± 0.05	–	–	–	–
170	Indene	95-13-6	1276	3	1050	1037	0.22 ± 0.05	–	–	–	–
171	Benzene, 1-methyl-3-propyl-	1074-43-7	1284	2.5	1054	1052	0.86 ± 0.15	–	–	–	–
172	Benzene, (1,1-dimethylethoxy)-	6669-13-2	1320	2.6	1072	1074	2.72 ± 0.04	–	–	–	–
173	Benzene, 1-ethyl-2,4-dimethyl-	874-41-9	1340	2.6	1082	1076	0.54 ± 0.07	–	–	–	–
174	Benzene, 4-ethyl-1,2-dimethyl-	934-80-5	1352	2.6	1088	1093	1.56 ± 0.06	–	–	–	–
175	Benzene, 1-ethenyl-3-ethyl-	7525-62-4	1352	2.7	1088	1084	2.30 ± 0.02	–	–	–	–
176	Benzenemethanol, a,a-dimethyl-	617-94-7	1352	3	1086	1080	0.54 ± 0.06	–	–	–	–
177	Benzene, 1-ethenyl-4-ethyl-	3454-07-7	1372	2.6	1093	1096	0.38 ± 0.08	–	–	–	–
178	4,7-Methano-1H-indene, octahydro-	6004-38-2	1388	2.5	1106	nf	0.72 ± 0.07	–	–	–	–
179	Benzaldehyde dimethyl acetal	1125-88-8	1396	2.9	1111	1080	0.46 ± 0.04	–	–	–	–
180	Naphthalene, decahydro-, cis-	493-01-6	1400	2.4	1114	1090	0.46 ± 0.07	–	–	–	–
181	Phenylethyl Alcohol	60-12-8	1408	3.2	1118	1136	0.45 ± 0.10	–	–	–	–
182	Naphthalene, decahydro-2-methyl-	2958-76-1	1420	2.3	1124	1122	1.93 ± 0.03	–	–	–	–
183	Benzene, 1-methyl-3-(1-methylethyl)-	535-77-3	1420	2.6	1124	1094	0.45 ± 0.03	–	–	–	–
184	Benzene, 1,2,4,5-tetramethyl-	95-93-2	1420	2.6	1124	1130	1.06 ± 0.3	1115	1435	nf	0.36 ± 0.09
185	1H-Indene, 2,3-dihydro-4-methyl-	824-22-6	1480	2.8	1148	1144	0.33 ± 0.03	–	–	–	–
186	Phenol, 3,5-dimethyl-	108-68-9	1500	2.9	1167	1178	0.52 ± 0.05	–	–	–	–
187	Naphthalene, 1,2,3,4-tetrahydro-	119-64-2	1508	2.9	1172	1154	2.81 ± 0.40	–	–	–	–
188	Benzoylhydrazine	613-94-5	1552	3	1194	1208	0.14 ± 0.02	–	–	–	–
189	Naphthalene	91-20-3	1552	3.2	1196	1194	1.37 ± 0.46	1566	1745	1740	0.94 ± 0.16	5^B
190	Benzene, 1,3-bis(1,1-dimethylethyl)-	1014-60-4	1652	2.4	1259	1249	0.70 ± 0.04	1105	1428	1431	0.41 ± 0.06
191	Benzaldehyde, 2,4,6-trimethyl-	487-68-3	1748	3	1318	1332	1.55 ± 0.11	–	–	–	–
192	Naphthalene, 2-methyl-	91-57-6	1748	3.1	1318	1329	1.53 ± 0.08	–	–	–	–	20^B
193	Naphthalene, 1,2,3,4-tetrahydro-1,1,6-trimethyl-	475-03-6	1840	2.8	1369	1402	0.31 ± 0.08	–	–	–	–
194	Phenol, 2,4-bis(1,1-dimethylethyl)-	96-76-4	2056	2.8	1509	1521	2.76 ± 0.31	–	–	–	–
195	2(4H)-Benzofuranone, 5,6,7,7a-tetrahydro-4,4,7a-trimethyl-	15356-74-8	2112	3.5	1554	1538	0.40 ± 0.03	–	–	–	–
196	1,2-Benzenedicarboxylic acid, bis(2-methylpropyl) ester	84-69-5	2520	2.5	1856	1863	0.97 ± 0.09	2471	2471	2452	1.02 ± 0.22
197	Dibutyl phthalate	84-74-2	2604	2.6	1918	1909	0.70 ± 0.01	–	–	–	–
Nitrogenous Compounds
198	Butanenitrile, 2-methyl-	18936-17-9	552	2.8	726	717	2.44 ± 0.24	–	–	–	–
199	Butanenitrile, 3-methyl-	625-28-5	568	2.8	733	730	2.35 ± 0.10	–	–	–	–
200	Pyrazine	290-37-9	596	3	745	747	13.66 ± 1.32	–	–	–	–
201	Pyridine	110-86-1	608	3	750	695	4.94 ± 0.69	–	–	–	–
202	1H-Pyrrole, 2-methyl-	636-41-9	804	3.1	834	860	0.73 ± 0.14	–	–	–	–
203	Oxime-, methoxy-phenyl-	0-00-0	932	1.7	894	886	114.83 ± 6.35	1598	1770	1405	51.58 ± 2.29
204	Ethosuximide	77-67-8	1656	3.3	1262	1193	0.79 ± 0.04	–	–	–	–
205	Formamide, N,N-dibutyl-	761-65-9	1736	2.8	1310	1319	0.39 ± 0.06	–	–	–	–
206	Caffeine	21399	2512	3.5	1851	1840	4.05 ± 0.35	–	–	–	–
Ethers
207	Furan, 2,3-dihydro-	1191-99-7	340	2.4	641	nf	2.82 ± 0.13	–	–	–	–
208	Furan, 2-methyl-	534-22-5	372	2.1	645	623	19.61 ± 0.63	–	–	–	–
209	Tetrahydrofuran	109-99-9	396	2.3	662	nf	27.74 ± 0.93	–	–	–	–
210	1,3-Dioxolane, 2-methyl-	497-26-7	424	2.4	670	676	13.39 ± 0.22	–	–	–	–
211	1,4-Dioxane	123-91-1	528	2.7	715	690	5.11 ± 0.02	–	–	–	–
212	n-Butyl ether	142-96-1	912	2.2	881	880	7.12 ± 1.79	–	–	–	–
213	6,8-Dioxabicyclo[3.2.1]octane, 1,5-dimethyl-, (1S)-	28401-39-0	1052	2.6	943	949	2.05 ± 0.10	–	–	–	–
214	Furan, 2-pentyl-	3777-69-3	1156	2.4	989	1000	5.51 ± 0.31	–	–	–	–
Sulfo compounds
215	Methyl thiolacetate	1534-08-3	508	2.64	707	675	4.08 ± 1.06
216	Disulfide, dimethyl	624-92-0	600	2.73	750	718	4.08 ± 0.67					12^A
217	2,4-Dithiapentane	1618-26-4	936	2.99	891	821	1.16 ± 0.49
218	Dimethyl Sulfoxide	67-68-5	980	3.51	911	772	1.14 ± 0.02	1316	1567	1596	0.81 ± 0.13
219	Dimethyl sulfone	67-71-0	988	0.6	913	915	5.08 ± 0.56	1779	1913	1912	1.57 ± 0.51

^aRI-cal, retention index of GC×GC-TOF MS with 5% phenyl polysilphenylene-siloxane GC column;

^bRef-ref, RI reported in the NIST Mass Search 2.0 and other references for 5% phenyl polysilphenylene-siloxane GC column or equivalents^[10,21,25];

^cRI-cal, retention index of GC×GC-TOF MS with DB-WAX GC column;

^dRef-ref, RI reported in the NIST Mass Search 2.0 and other references WAX GC column or equivalents^[22–24];

Values were expressed as mean ± standard deviation (n = 3);

^AThreshold values (ng/g) in water by Bendall;^[10]

^BOdor threshold values (μg/L) in water by Yang et al.^[30]

^CThreshold values (μg/L in water by Honkanen et al.^[31]]

TABLE 2 Number and quantitative results of volatile compounds in squid samples by GC-MS and GC×GC-TOF MS analysis

Compounds	GC×GC-MS			GC-MS
	N	RC (%)	Concentration (ng/ mL milk)	N	RC (%)	Concentration (ng/ mL milk)
Aliphatic Hydrocarbons	69	26.66	617.08	5	1.13	5.32
Alcohols	14	6.39	147.96	4	1.17	5.5
Aldehydes	14	5.56	128.64	3	1.32	6.2
Ketones	28	20.90	483.88	10	34.45	161.67
Acids	14	18.67	432.24	11	38.29	179.71
Esters	16	4.19	97.03	5	4.62	21.67
Aromatic Hydrocarbons	42	7.12	164.92	11	7.51	35.25
Nitrogenous Compounds	9	6.23	144.18	1	10.99	51.58
Ethers	8	3.60	83.35	–	–	–
Sulfo compounds	5	0.67	15.54	2	0.51	2.38
Total	219		2314.82	52		469.28

N, number of compounds; RC, relative content of peak area.

Acids are derived from lipolysis, degradation of lactose and amino acids, which are responsible for perceptible rancid flavor in milk.^[8,²⁰,²⁶,^27] In the milk, carboxylic acids are not only crucial aroma themselves, but also are important precursors of other compounds, including methyl ketones, alcohols, aldehydes, and esters.^[28] Acetic acid, butanoic acid, hexanoic acid, and octanoic acid were also found in 2% butterfat milk.^[29] In addition, decenoic acid, tetradecanoic acid, hexadecanoic acid, oleic acid, and lauric acid were reported mainly the acid compounds in pulsed electric field treated milk.^[7] Similarly, aromatic hydrocarbons have also been widely identified in milk.^[11,¹³,^15] For example benzene, toluene, ethylbenzene, and indene have been found by Wang et al. in reconstituted milk and reduced-fat cheese.^[8] Bendall suggested that phenol compounds impart woody, smoky, and burnt aromas to milk and have been considered as the agents responsible for the smell of cow urine.^[10] Some of the aromatic compounds are derived from the Strecker reaction, e.g., α-oxidation of phenyl acetaldehyde is responsible for benzaldehyde production that imparts aromatic notes of bitter almond to dairy products.^[20]

Following the predominant groups are ketones, aliphatic hydrocarbons, and esters, in number of compounds. Among the ten identified ketone components, the top four compounds were acetone (87.44 ng/mL milk), 2-Butanone (32.55 ng/mL milk), 2-Heptanone (22.19 ng/mL milk), and 2-Pentanone (4.33 ng/mL milk; Table 1). The aliphatic hydrocarbons and esters comprised five compounds each. The former included decane, undecane, dodecane, tridecane, and tetradecane; and the latter comprised ethyl acetate, butanoic acid methyl ester, acetic acid butyl ester, propanoic acid butyl ester, and butanoic acid butyl ester. Methyl ketones are the main ketone group of volatile compounds in milk.^[3,¹¹,^15] It can be supposed that acetone and butanone originated from bovine metabolism.^[11] Both 2-pentanone and 2-heptanone are partly thermally-induced compounds formed upon thermal decarboxylation of β-ketoacids or β-oxidation of fatty acids, followed by decarboxylation.^[3,20]

Aliphatic hydrocarbons have a weak odor, are generally considered to have less impact on the overall milk flavor because of their high aroma thresholds.^[20] The important contributions of the ester compounds to food aroma are undisputed: esters with low carbon atoms are highly volatile at ambient temperatures and the perception thresholds are ten times lower than their alcohol precursors.^[20] In the present study, ethyl acetate, due to its high amount, probably contribute acesent/sour flavor to the milk, which originates from the action of lactic acid bacteria.

In addition, four alcohols (1-Butanol, 2-ethoxy-Ethanol, 2-ethyl-1-Hexanol, and 1-Octanol), three aldehydes (hexanal, octanal, and nonanal), two sulfo compounds and one nitrogenous compound (methoxy-phenyl-Oxime) were detected. Alcohols, which are responsible for the strong and exciting flavors in dairy product, could be formed by reduction from the aldehydes,^[20] amino acid metabolism or fermentation of lactose.^[8] Most volatile alcohols have little impact on food flavors because of their high odor threshold, unless they are unsaturated or present at high concentrations.^[20] Acetaldehydes have been reported in heated milk as a product of light induced oxidation in milk or PET degradation.^[15,^32] Hexanal, a grass aroma, is the main aldehyde detected in our GC-MS analysis, which is a typical volatile from oxidized linoleic acid,^[11] while octanal and nonanal were present in smaller proportions. These three aldehydes known to contribute off-flavors to milk were also found in stale flavor volatiles from the headspace of UHT milk.^[13,^29] There is little information on methoxy-phenyl-oxime, but the compound has also been found in in ultrapasteurized milk packaged in polyethylene terephthalate containers and in reconstituted milk and reduced-fat cheese.^[3,^8] Aldehydes could provide significant aromas, either pleasant or rancid, to foodstuffs, and their odor threshold values are usually lower than those of alcohols.^[25] Thus, even trace amounts of aldehydes might override the flavor effect of some other substances.^[20] Sulfur containing molecules are responsible for the “cooked” off-flavor developed during high temperature pasteurization.^[7,^12] In this study, only dimethyl sulfone and dimethyl sulfoxide were detected by GC-MS in microfiltered pasteurized milk and their contents were 1.57 and 0.81 ng/mL milk, respectively. This may attribute to the mass spectrometric detectors or the SPME technique being insufficiently sensitive to detect other sulfur compounds with the low concentrations.^[2,7] In the future work, the extraction parameters and detectors would be optimized to improve the sensitivity.

Table 2 also shows that the contents of eight volatile compounds in milk in decreasing order were acids > ketones > nitrogenous compounds > aromatic hydrocarbons > esters > aldehydes > alcohols > aliphatic hydrocarbons > sulfo compounds. Among the 52 identified volatile components, the dominant five volatiles were acetone (87.44 ng/mL milk), octanoic acid (66.94 ng/mL milk), methoxy-phenyl-oxime (51.58 ng/mL milk), hexanoic acid (47.67 ng/mL milk), and n-Decanoic acid (42.56 ng/mL milk).

As reported previously, when using GC-MS methods, many fewer (11 to 80) volatile compounds in different milk products could be identified (Table 2).^[3,⁶,⁷,⁹,^11–15,^33] The differences observed might reflect the different volatile extraction methodologies and milk processing methods used, chemical compositions in the milk, biological factors, and/or geographical origin of the cows.

SPME-GC×GC-TOFMS Analysis

Group-type classification

A 3D partial plot of milk sample is presented in Fig. 1, which indicated that the volatile compounds in the milk were very complicated. In the TIC, the analytes in the contour plots could be divided into five main classes (Fig. 2): aliphatic hydrocarbons, alcohols, aromatic compounds, ketones, and aldehydes. Some other volatiles, e.g., esters and nitrogenous compounds, could also be detected, but they overlapped each other or their levels were low, and are thus, not emphasized in the figures.

FIGURE 1 3D partial surface plot of milk sample.

FIGURE 2 Structurally ordered color plot of main compound classes of flavor volatiles of milk.

Identification

The analysis of the milk samples requires that each peak is identified based on its ordered retention time and MS confirmation. The total peak areas caused by solvent elution and column bleeding were removed using the features of ChromaToF, and the peak areas of analytes were converted into the new relative area, which represented 100%. The TIC was used to set up the minimum relative peak area and similarity criterion to isolate the most abundant analytes. The first class, with relative peak area greater than 1%, comprised the top 21 compounds (Table 3). The second (47) and third class (84) were compounds with relative peak areas of more than 0.5 and 0.2%, respectively (Table 3). According to their chemical structures, all compounds could be divided into nine groups: aliphatic hydrocarbons, aromatic hydrocarbons, esters, ketones, alcohols, nitrogenous compounds, aldehydes, ethers, and acids. The total relative percentages of peak areas and the numbers of compounds in each group are shown in Table 2. Among the groups, aliphatic compounds presented the highest number, followed by aromatic hydrocarbons, ketones, esters, aldehydes/alcohols/acids, nitrogenous compounds, ethers, and sulfo compounds. Whereas, the top five groups with the highest relative area percentages were aliphatic compounds, ketones, acids, aromatic hydrocarbons, and alcohols.

TABLE 3 Number of peaks based on minimum peak areas

Criteria	Number of peaks detected
Total number of detected peaks (between brackets: relative area)	215 (100%)
Peak area > 1%	22 (62.49%)
Peak area > 0.5%	47 (79.19%)
Peak area > 0.2%	84 (91.04%)

Milk volatiles profile

Table 1 shows that a total of 219 volatile compounds were identified by GC×GC-TOFMS, which were detected in the first dimension range of 320–2604 s and in a second dimension range of 0.1–3.8 s, and the respective RIs were calculated according to the ChromaTOF software system. An intensive literature search was conducted to obtain the RI values for the compounds detected in the present study (Table 1, RI_Ref). The most intensive peaks, called base peaks, were used to calculate the RIs. Compared with the RIs reported in the literature (Table 1), a maximum difference of 30–40 was found for the absolute RI (|RI_cal-RI_Ref|) in this study, and this variation can be considered reasonable.^[17] Compound types, numbers and contents correspond to those presented in Tables 1 and 2. The analysis by GC×GC-TOFMS allowed the authors to identify about four times more compounds than were detected by GC-MS. In particular, eight ethers identified by GC×GC-TOFMS were not detected by GC-MS in milk. In addition, the analysis by GC×GC-TOFMS allowed an additional detection of: aliphatic hydrocarbons (64), aromatic hydrocarbons (31), ketones (18), esters (11), aldehydes (11), alcohols (10), nitrogenous compounds (8), sulfo compounds (3), and acids (3), compared with GC-MS. Moreover, the contents of nine volatile compounds in milk in decreasing order were aliphatic hydrocarbons > ketones > acids > aromatic hydrocarbons > alcohols > nitrogenous compounds > aldehydes >esters > ethers > sulfo compounds. As shown in Table 2, the concentrations of volatiles detected using GC×GC-TOFMS were about 4.93-fold higher than those determined by GC-MS. Among the 214 identified volatile components, the dominant five volatiles were hexanoic acid (193.57 ng/mL milk), methoxy-phenyl-oxime (114.83 ng/mL milk), octanoic acid (109.38 ng/mL milk), 4,5-dimethyl-1-hexene (101.48 ng/mL milk), and 2-Pentanone (99.74 ng/mL milk).

Compared with the GC system, the greater capacity of GC×GC to identify larger number and amounts of compounds lies in its second dimension column, as well as the spectral deconvolution function for the co-eluted compounds in both the first and second dimension. As shown in Fig. 3a, the three compounds that overlapped in the first dimension time at 1176 s were separated by the second dimension column, and identified as decane, 1,1,3,3,5-Pentamethylcyclohexane, (S)-2-Octanol, respectively (Fig. 3b). Their second dimension retention times were 2.010, 2.100, and 2.320 s, respectively, showing a great difference. Thus, they could be effectively separated. However, only decane was detected by GC-MS, indicating that these three compounds co-eluted in GC-MS analysis, which led to inaccurate quantitative analysis of aliphatic hydrocarbons and alcohols. Figure 4a showed that three chromatographic peaks overlapped in the first and second dimension time. The spectral deconvolution based on mass spectra differences isolated the peaks and identified them as 2,2,4,6,6-pentamethyl-heptane, 4-ethyl-2-Octene and1-Decyne (Fig 4b). This indicated that GC×GC-TOFMS has much higher potential to isolate and identify compounds than the GC-MS detection method.

FIGURE 3 An example of a more powerful separation ability of 2D GC compared to 1D GC (A) Separation of three compounds by the second dimension column at 1176 s (¹tR) found in volatiles of milk: (1) Decane, 2.010 s (²tR); (2) 1,1,3,3,5-Pentamethylcyclohexane, 2.100 s (²tR); (3) 2-Octanol, (S)-, 2.320 s (²tR); (B) Deconvoluted mass spectra of compounds in (A). The three compounds were separated by the second dimensional column, and would have co-eluted on the first dimensional column.

FIGURE 4 (A) Modulated peaks of three compounds found in flavor volatiles of milk: (1) red line, Heptane, 2,2,4,6,6-pentamethyl-, m/z 57; (2) green line, 2-Octene, 4-ethyl-, m/z 55; (3) blue line, 1-Decyne, m/z 67. (B) Deconvoluted mass spectra of compounds in (A).

According to the data available in the literature,^[1,^3–15,²¹,²⁵,²⁶,^33–35] among the 219 volatiles detected by GC×GC-TOFMS, 107 are reported for the first time in milk: aliphatic hydrocarbons (15), aromatic hydrocarbons (30), esters (19), ketones (9), alcohols (14), nitrogenous compounds (12), aldehydes (2), acids (3), and furans (3). These compounds have already been detected in other food, and some of them have been reported to have sensory properties and characteristic aromas. For example, 4,5-dimethyl-1-Hexene was detected as a major compound, comprising approximately 16.4% of the total aliphatic hydrocarbons, which has fruits or with fragrances.^[20] 1,5-Hexadien-3-ol, (Z)-2-decenal (slightly fatty, floral, green) and 2-pentylfuran (are beany, grassy, and licorice-like tastes) are also well-known and common aroma compound).^[36] In addition, isobutyl acetate was detected in trace amounts in milk and its aroma descriptor related to the characteristic flavor of pear.^[20]

CONCLUSION

The qualitative analysis of milk volatiles using GC×GC-TOFMS was reported for the first time. The present results showed that GC×GC-TOFMS improved the resolution and separation efficiency of highly complex analytes like milk. It not only separated the 52 compounds that coelute in conventional GC-MS, but also allowed the identification of compounds that were not previously detected. The milk samples were found to contain 219 volatiles by GC×GC-TOFMS, including aromatic hydrocarbons, aliphatic hydrocarbons, esters, aldehydes, alcohols, nitrogenous compounds, ketones, acids, ethers, and sulfo compounds. One hundred seven of these compounds were reported for the first time in milk compared with previously published results from GC-MS. Both the number and concentration of compounds identified using GC×GC-TOFMS were higher than those detected by GC-MS analysis for milk volatiles. Overall, SPME-GC×GC-TOFMS was able to extract, separate, and identify the volatile compounds from the milk sample; but it still lacks information on the key aroma compounds. Thus, how to separate and identify the key aroma compounds will be addressed in the future.

FUNDING

This research was supported by grants from the Open Project Program of State Key Laboratory of Dairy Biotechnology, Bright Dairy and Food Co. Ltd. (SKLDB2013-02), the Shanghai Minhang District Commission of Science and Technology(2013MH088), the National Science and Technology Pillar Program during the 12th Five-year Plan Period (No. 2013BAD18B02), and the Agricultural Science and Technology Achievements Transformation Fund (2012GB2CO00141).

ACKNOWLEDGMENTS

A special thanks to SJTU-Instrumental Analysis Center for expert assistance with the GC-MS and GC×GC-TOFMS experiments.

Additional information

Funding

This research was supported by grants from the Open Project Program of State Key Laboratory of Dairy Biotechnology, Bright Dairy and Food Co. Ltd. (SKLDB2013-02), the Shanghai Minhang District Commission of Science and Technology(2013MH088), the National Science and Technology Pillar Program during the 12th Five-year Plan Period (No. 2013BAD18B02), and the Agricultural Science and Technology Achievements Transformation Fund (2012GB2CO00141).